Project Management Journal

P ROJECT M ANAGEMENT J OURNAL

Research Quarterly, Volume 37, Number 3 A UGUST 2006 Special PMI Research Conference 2006 Edition 5 16 26 37 49

From Nobel Prize An Empirical On the Novelty The Perceived Value Mechanisms for to Project Investigation of the Dimension and Potential Inter-project Management: Sponsor’s Role in in Project Contribution of Integration– Getting Project Initiation Management Project Management Empirical Analysis Risks Right Timothy J. Kloppenborg, Klaus Brockhoff Practices to Project in Program Context Deborah Tesch, Bent Flyvbjerg Chris Manolis, and Success Perttu Dietrich Mark Heitkamp Claude Besner and Brian Hobbs

62 74 87 98 111

Longitudinal Developing Project-Based A Theoretical New Possibilities Analysis of Organizational Management as an Framework for for Project Project Project Management Organizational Aligning Project Management Theory: Innovation: Drivers, Management Capability: Management A Critical Engagement Changes, and Maturity Theory and Practice Benefits of with Business Svetlana Cicmil and Damian Hodgson Mark Mullaly Lynn Crawford Adopting Project- Strategy Based Management Dragan Z. Milosevic and Miia Martinsuo, Sabin Srivannaboon Nicole Hensman, Karlos Artto, Jaakko Kujala, and Ali Jaafari

Making Project Management Indispensable for Business Results™

Published by

P ROJECT M ANAGEMENT J OURNAL

The Professional Research Journal of the Project Management Institute

Volume 37, Number 3 AUGUST 2006

3 From the Editor Christophe N. Bredillet, PhD, MBA, Ingénieur EC Lille

PAPERS

5 FROM NOBEL PRIZE TO PROJECT MANAGEMENT: GETTING RISKS RIGHT Bent Flyvbjerg

16 AN EMPIRICAL INVESTIGATION OF THE SPONSOR’S ROLE IN PROJECT INITIATION Timothy J. Kloppenborg, Deborah Tesch, Chris Manolis, and Mark Heitkamp

26 ON THE NOVELTY DIMENSION IN PROJECT MANAGEMENT Klaus Brockhoff

37 THE PERCEIVED VALUE AND POTENTIAL CONTRIBUTION OF PROJECT MANAGEMENT PRACTICES TO PROJECT SUCCESS Claude Besner and Brian Hobbs

49 MECHANISMS FOR INTER-PROJECT INTEGRATION–EMPIRICAL ANALYSIS IN PROGRAM CONTEXT Perttu Dietrich

62 LONGITUDINAL ANALYSIS OF PROJECT MANAGEMENT MATURITY Mark Mullaly

74 DEVELOPING ORGANIZATIONAL PROJECT MANAGEMENT CAPABILITY: THEORY AND PRACTICE Lynn Crawford

87 PROJECT-BASED MANAGEMENT AS AN ORGANIZATIONAL INNOVATION: DRIVERS, CHANGES, AND BENEFITS OF ADOPTING PROJECT-BASED MANAGEMENT Miia Martinsuo, Nicole Hensman, Karlos Artto, Jaakko Kujala, and Ali Jaafari

98 A THEORETICAL FRAMEWORK FOR ALIGNING PROJECT MANAGEMENT WITH BUSINESS STRATEGY Dragan Z. Milosevic and Sabin Srivannaboon

111 NEW POSSIBILITIES FOR PROJECT MANAGEMENT THEORY: A CRITICAL ENGAGEMENT Svetlana Cicmil and Damian Hodgson

123 Notes for Authors

124 Index of Advertisers PROJECT MANAGEMENT JOURNAL

EDITORIAL REVIEW BOARD Editor Project Management Institute Global Operations Rick Bilbro, The Innova Group, Inc.; Christophe N. Bredillet, PhD, MBA, Center, Four Campus Blvd, Newtown Square, PA 19073- Ingénieur EC Lille 3299 U.S.A.; tel: +1-610-356-4600; fax +1-610-356-4647 Tomas Blomquist, Umeå University, Sweden; [email protected]; mailto: [email protected] Pierre Bonnal, CERN Accelerator and Beams Dept.; PMI PUBLISHING STAFF Web site: www.pmi.org David Bonyuet, PMP, Delta Search Labs; PMI Regional Service Centre: Europe - Middle East David Christensen, Cedar City, UT; Interim Publisher - Africa (EMEA), 300, Avenue de Tervueren Donn Greenberg; [email protected] B-1150 Brussels, Belgium; tel. +32-2-743 15 73; fax +32- Dan H. Cooprider, Creative People Management; 2-743 15 50; e-mail: [email protected] Steven V. DelGrosso, IBM; Contributing Editor mailto:[email protected] Dan Goldfischer; [email protected] J. Davidson Frame, University of Mgmt. & Technology ; PMI Asia Pacific Service Centre, #03-01, Rex House 73 Bukit Timah Road; Singapore 229832; tel. + 65 6330 Gary C. Humphreys, Humphreys & Associates; Project Management Journal Associate 6733; fax. +65 6336 2263; E-mail: [email protected] Natasha Pollard; [email protected] Peter Kapsales, Belcore; mailto:[email protected] Young Hoon Kwak, The George Washington University; Publications Planner PMI Representative Offices, Beijing, China Alexander Laufer, Technion-Israel Inst. of Technology; Barbara Walsh; [email protected] Washington, D.C. William Leban, PMP, DeVry University; Bookstore EDITORIAL & ADVERTISING SERVICES Rolf A. Lundin, Jonkoping University; [email protected] Address manuscripts and other editorial submissions, Fred Manzer, PMP, Center for Systems Management; advertising and mailing list rental inquiries, and requests General e-mail: Peter Morris, University College, London; for reprints/bulk copies/reprint permission to: Project [email protected] Management Institute, Publishing Department, Four Ralf Müller, Umeå University, Sweden; Journal in print: Campus Boulevard, Newtown Square, PA 19073-3299 Michael Okrent, Agilent Technologies Inc.; [email protected] USA. tel: +1-610-356-4600, fax: +1-610-356-4647; e-mail: Blaize Horner Reich, University of British Columbia|Okanagan; [email protected] Paul B. Robinson, The Eastern Management Group; Book Review Editor Unless otherwise specified, all letters and articles sent to the PMI Publishing Department are assumed for Arnold M. Ruskin, PMP, Claremont Consulting Group; Kenneth H. Rose, PMP publication and become the copyright property of PMI if Richard L. Sinatra, PMP, Potomac, MD; Contributing Editor published. PMI is not responsible for loss, damage, or James Snyder, Springfield, PA; Lisa M. Fisher injury to unsolicited manuscripts or other material. Anders Soderholm, Umea University; READER SERVICES Paul Solomon, PMP, B-2 Earned Value Photocopies. Project Management Journal has been Management Systems; Boulevard, Newtown Square, PA 19073-3299 USA; tel: +1- registered with the Copyright Clearance Center, Inc. Charles J. Teplitz, University of San Diego; Consent is given for copying of articles for personal or 610-356-4600, fax: +1-610-356-4647. Janice Thomas, Athabasca University; internal use, or for the personal use of specific clients. The mission of the Journal is to provide information This consent is given on the condition that the copier Zeljko M. Torbica, PMP, Florida advancing the state of the art of the knowledge of project International University; pays through the Center the per copy fee stated in the management. The Journal is devoted to both theory and code on the first page of each article for copying beyond Terry Williams, University of Strathclyde; practice in the field of project management. Authors are that permitted by Sections 107 or 108 of the U.S. Graham M. Winch, The University of Manchester. encouraged to submit original manuscripts that are derived Copyright Law. The appropriate fee should be forwarded from research-oriented studies as well as practitioner case with a copy of the first page of the article to the EDITORIAL ADVISORY BOARD studies. All articles in the Journal are the views of the Copyright Clearance Center, Inc., 222 Rosewood Drive, Frank T. Anbari, PMP, The George authors and are not necessarily those of PMI. Subscription Danvers, MA 01923 USA (tel: +1-508-750-8400; fax: Washington University; rate for members is $14 U.S. per year and is included in the +1-508-750-4744). The fee indicated on the first page of an article in this issue will apply retroactively to all Karlos Artto, Helsinki University of Technology annual dues. articles published in the journal, regardless of the year of (HUT), Finland; Claims for undelivered copies must be made no later publication. This consent does not extend to other kinds David Cleland, University of Pittsburgh; than two months following month of publication. The of copying, such as for general distribution, resale, Lynn Crawford, University of Technology, Sydney; publisher will supply missing copies when losses have been advertising and promotion purposes, or for creating new Brian Hobbs, Université du Québec à Montréal; sustained in transit and when the reserve stock will permit. collective works. Special written permission must be Toshihiko Kinoshita, Waseda University; PMI is a nonprofit professional organization whose obtained from the publisher for such copying. mission is to serve the professional interests of its Permissions. Requests to reprint articles published Mark E. Nissen, Naval Postgraduate School; collective membership by: advancing the state of the art in Project Management Journal must be made in writing to Asbjørn Rolstadås, Norwegian University of in the leadership and practice of managing projects and the publisher. Science and Technology; programs; fostering professionalism in the management of Reprints. Individual articles may be purchased through the Knowledge & Wisdom Center Dennis P. Slevin, University of Pittsburgh. projects; and advocating acceptance of project (www.pmi.org/k&wc) management as a profession and discipline. Membership PUBLICATION & MEMBERSHIP Glossy Reprints. Requests for glossy reprints of in PMI is open to all at an annual dues of $119 U.S. For The Project Management Journal (ISSN 8756-9728/03) individual articles in quantities of 100 or more can be sent information on PMI programs and membership, or is published quarterly (March, June, September, December) to [email protected]. to report change of address or problems with your by the Project Management Institute. The Journal is printed Bulk Copies of Current Issues. Copies of the subscription, contact: in the USA by Cadmus Magazine, Richmond, VA. current Project Management Journal can be obtained in quantities of 25 or more. Orders must be placed 40 days Periodical postage paid at Newtown Square, PA 19073 USA prior to date of issue. The cost is $5.50 per copy plus and at additional mailing offices. Canadian agreement shipping. #40030957. Postmaster: Send address changes to Project Back Issues. Back issues are available on request at Management Institute, Headquarters, Four Campus $20 each. Call +1-610-356-4600 for availability.

© 2006 Project Management Institute, Inc. All rights reserved. “PMI” is a trade and service mark registered in the United States and other nations; “PMP” and the PMP logo are registered certification marks in the United States and other nations; “PMBOK” is a trademark registered in the United States and other nations; and the PMI logo, “PM Network”, “Project Management Journal”, “PMI Today”, and “Building professionalism in project management.” are trademarks of the Project Management Institute, Inc. F ROM THE EDITORS Christophe Bredillet, PhD, DSc, MBA, Ingénieur EC Lille, CPD IPMA Level A, PRINCE2 Practitioner, CCE, ESC Lille, France Dragan Milosevic, PhD, MBA, BSCE, PMP, Portland State University, USA Janice Thomas, PhD, MBA, BSc, Athabasca University, Canada Terry Williams, PhD, MSc, MA, FORS CMath, PMP, University of Southampton, United Kingdom

elcome to this special edition of the Project decades” (Flyvbjerg, this edition). Building from the W Management Journal featuring some of the findings of a major research project exploring mega most interesting papers presented at PMI’s projects and risk in the civil engineering field, Biannual Research Conference held in Montreal, Flyvbjerg outlines the idea and use of reference class Canada earlier this year. forecasting, making a solid argument for its applica- Papers were selected for their unique contribu- bility and in fact necessity to improve project out- tions to the project management research literature comes in all industries. This research makes a by a panel of four respected project management significant contribution to the project management researchers, who also served as co-editors for this literature by providing empirical support for Special PMI Research Conference 2006 Edition. improving how we develop and use estimates. These academics reviewed each of the papers Executive level support is widely recognized as prepared for the research conference, and their rat- playing an important role in achieving project suc- ings by the conference reviewers, and selected those cess and yet very little research explores exactly what for publication in this special edition. these project executives do to facilitate project suc- Papers included in this special edition reflect cess. Kloppenborg, Tesch, Manolis, and Heitkamp the global growth in project management research: present “An Empirical Investigation of the Sponsor’s four papers are from North American scholars, five Role in Project Initiation,” which examines and clas- are from Europeans and one is from an Australian. sifies behaviors associated with the role of the proj- Some of these papers come from established schol- ect sponsor; “the executive with the fiscal authority, ars; others are from relatively newcomers to the political clout, and personal commitment to see a research world. Several provide empirically based project through.” This study goes beyond identifying explorations of the state of project management in project sponsor behaviors, empirically validating today’s organizations. Others provide theoretical and then testing the association of these behaviors discussions designed to shape project management with project success outcomes in specific situations. thinking and research. Some are provocative. All are Project classification and differentiation is rigorously grounded. Together these papers present another important aspect of improving project man- unique insights into the state of project manage- agement by ensuring that the right tools and tech- ment research and practice to date. niques are applied to each and every project. The first papers in the special edition explore Brockhoff contributes a paper “On the Novelty important longstanding issues in single project Dimension of Project Management,” introducing management (Flyvbjerg; Kloppenborg et al.; results from German studies that model dimensions Brockhoff; Besner and Hobbs). The next set of that differentiate projects that are not yet reflected in papers investigates multiproject management and the practice or literature on project management. the organizational impact of project management These studies conclude that although more formal (Dietrich; Mullaly; Crawford; Martinsuo et al.). The project management methods are adequate for less final two papers develop new theoretical approach- novel projects, the characteristics of project man- es to our study, understanding, and practice of proj- agers may be more important than project manage- ect management (Milosevic and Srivannaboon; ment methods for achieving success in highly Cicmil and Hodgson). We hope you enjoy them. innovative projects. Even more importantly we trust and expect that they Addressing the issue of the tools and practices will make you think. used in project management, Besner and Hobbs Practitioners continually request more research explore the “The Perceived Value and Potential on project estimating as “Forecasts of cost, demand, Contribution of Project Management Practices to and other impacts of planned projects have Project Success.” Survey responses from 753 proj- remained constantly and remarkably inaccurate for ect management practitioners provide interesting

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 3 conclusions as to which project management practices drives an organization to implement project manage- are most valued for their impact on project success. ment. The authors explore what changes are imple- Organizational learning is recognized as one of the most mented in such an initiative and what the resulting valuable tools for its perceived ability to improve project benefits are to the organization of embarking on such a performance. In contrast to Brockhoff’s paper, this paper change. Interesting conclusions as to the primary drivers concludes that practitioners perceive many project man- of project management initiatives within organizations agement tools are available for large and more complex are presented. projects, while tools and techniques for small, less novel Milosevic and Srivannaboon, in “A Theoretical and internal projects are lacking. Framework for Aligning Project Management With Moving from a focus on single project management, Business Strategy,” present an empirically based theoreti- Dietrich tackles important issues surrounding multipro- cal framework that highlights the impact of business ject programs in their paper entitled “Mechanisms for strategy on project management (and vice-versa) and the Inter-Project Integration—Empirical Analysis in Program mechanisms used to strengthen that alignment. In look- Context.” Extending the contingency models of projects ing at the alignment issue as a two-way influence between to look at their variations across programs, this paper project management and strategy, they provide insight describes an exploratory study of four independent case into the complex processes needed to accomplish this studies. These cases were designed to explore how per- alignment in organizations and present a potential ceived uncertainty and structural complexity affect the framework for integrating and supporting future research perceived importance of different integration mecha- in this area. nisms that are used in multiproject programs. Finally, Cicmil and Hodgson, in the paper “New Many models for assessing project management Possibilities for Project Management Theory: A Critical maturity have arisen over the last 5–10 years; however, Engagement,” introduce alternative theoretical approach- there is very little longitudinal data in the public domain es to the study of projects to problematise what is accept- to provide an understanding of the evolving state of prac- ed as known about projects and project management. tice. Mullaly’s paper, “Longitudinal Study of Project They very nicely summarize the issues and concerns of Management Maturity,” explores the theory underlying both academics and practitioners with the research and such assessments, the evolution of maturity testing in a literature on project management to date, providing project management context, and results from one matu- insights into why projects merit serious research attention rity assessment tool over a period of six years. This paper and why practically they are rising in popularity and provides a longitudinal analysis of the changes in project importance in contemporary organizations. In so doing, management capabilities over time that will allow practi- they broaden the research agenda for developing aca- tioners, researchers, and organizations to understand the demically important and practically interesting research. impact of project management maturity initiatives. Together, these papers, first, provide a provocative, In a similar vein, Crawford’s paper, “Developing diverse yet solid exploration of the state of project man- Organizational Project Management Capability Theory agement in today’s organizations, second, reflect direc- and Practice,” explores the development of project man- tions of the project management research. In other agement capabilities over time in one organization as jux- words, the collected papers provide a comprehensive taposed with the evolution organizational project cross-section of the issues that practically face organiza- management capability as represented in project man- tions and that theoretically confront researchers. It is agement literature and standards. Using a discourse both a summation of the work that has been conducted analysis approach and exploring both company reports to date, and an exciting basis for future research and on movement toward increasing capability and the exploration into how organizations are implementing, research and standards literature on how to do this, adopting, and adapting project management. We offer it Crawford compares and contrasts espoused theory with in the hope that it is received with the same excitement theories in use over time. This paper raises interesting and enthusiasm with which it was assembled. questions about the role literature and standards play in following or leading project management practice. Christophe N. Bredillet Rather than focusing on the specific levels of change Dragan Milosevic to project management maturity or capability, Martinsuo Janice Thomas and colleagues investigate “Project-Based Management as Terry Williams an Organizational Innovation: Drivers, Changes and Benefits of Adopting Project-Based Management.” Using institutional theory as a theoretical framework, this study employs survey-based data collection to understand what

4 AUGUST 2006 PROJECT MANAGEMENT JOURNAL FROM NOBEL PRIZE TO PROJECT MANAGEMENT: GETTING RISKS RIGHT

BENT FLYVBJERG, Aalborg University, Denmark

The American Planning Association Endorses Reference Class Forecasting ABSTRACT n April 2005, based on a study of inaccuracy in demand forecasts for public works projects by Flyvbjerg, Holm, and Buhl (2005), the American Planning A major source of risk in project manage- Association (APA) officially endorsed a promising new forecasting method ment is inaccurate forecasts of project I costs, demand, and other impacts. The called “reference class forecasting” and made the strong recommendation that paper presents a promising new planners should never rely solely on conventional forecasting techniques when approach to mitigating such risk based making forecasts: on theories of decision-making under APA encourages planners to use reference class forecasting in addition to uncertainty, which won the 2002 Nobel traditional methods as a way to improve accuracy. The reference class fore- Prize in economics. First, the paper documents inaccuracy and risk in project man- casting method is beneficial for non-routine projects... Planners should never agement. Second, it explains inaccuracy rely solely on civil engineering technology as a way to generate project fore- in terms of optimism bias and strategic casts (American Planning Association, 2005). misrepresentation. Third, the theoretical Reference class forecasting is based on theories of decision-making under basis is presented for a promising new uncertainty that won Princeton psychologist Daniel Kahneman the Nobel prize in method called “reference class forecasting,” which achieves accuracy by basing economics in 2002 (Kahneman, 1994; Kahneman & Tversky, 1979a; 1979b). forecasts on actual performance in a ref- Reference class forecasting promises more accuracy in forecasts by taking a so- erence class of comparable projects and called “outside view” on prospects being forecasted, while conventional forecast- thereby bypassing both optimism bias ing takes an inside view. The outside view on a given project is based on and strategic misrepresentation. Fourth, knowledge about actual performance in a reference class of comparable projects. the paper presents the first instance of practical reference class forecasting, Where Flyvbjerg, Holm, and Buhl (2005) briefly outlined the idea of reference which concerns cost forecasts for large class forecasting, this paper presents the first instance of reference class forecasting transportation infrastructure projects. in practical project management. The emphasis will be on transportation project Finally, potentials for and barriers to ref- management, because this is where the first instance of reference class forecasting erence class forecasting are assessed. occurred. It should be mentioned at the outset, however, that comparative research Keywords: risk management; project shows that the problems, causes, and cures identified for transportation apply to forecasting; forecast models a wide range of other project types, including concert halls, museums, sports arenas, exhibit and convention centers, urban renewal, power plants, dams, water ©2006 by the Project Management Institute projects, IT systems, oil and gas extraction projects, aerospace projects, new pro- Vol. 37, No. 3, 5-15, ISSN 8756-9728/03 duction plants, and the development of new products and new markets (Altshuler & Luberoff, 2003; Flyvbjerg, 2005; Flyvbjerg, Bruzelius, & Rothengatter, 2003, pp. 18–19; Flyvbjerg, Holm, & Buhl, 2002, p. 286).

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 5 Inaccuracy in Forecasts ture investments, the consequence is and the use of inappropriate forecast- Forecasts of cost, demand, and other inaccuracy to the second degree. ing models (Vanston & Vanston, 2004, impacts of planned projects have Benefit-cost ratios are often wrong, not p. 33). However, when such explana- remained constantly and remarkably only by a few percent but by several fac- tions are put to empirical test, they do inaccurate for decades. No improve- tors. This is especially the case for rail not account well for the available data. ment in forecasting accuracy seems to projects (Flyvbjerg, Bruzelius, & First, if technical explanations were have taken place, despite all claims of Rothengatter, 2003, pp. 37–41). As a valid, one would expect the distribu- improved forecasting models, better consequence, estimates of viability are tion of inaccuracies to be normal or data, etc. (Flyvbjerg, Bruzelius, & often misleading, as are socioeconomic near-normal with an average near zero. Rothengatter, 2003; Flyvbjerg, Holm, & and environmental appraisals, the accura- Actual distributions of inaccuracies are Buhl, 2002; 2005). For transportation cy of which are all heavily dependent on consistently and significantly non-nor- infrastructure projects, inaccuracy in demand and cost forecasts. These results mal with averages that are significantly cost forecasts in constant prices is on point to a significant problem in trans- different from zero. Thus the problem average 44.7% for rail, 33.8% for portation project management: More is bias and not inaccuracy as such. bridges and tunnels, and 20.4% for often than not, the information that man- Second, if imperfect data and models roads (see Table 1).1 For the 70-year agers use to decide whether to invest in were main explanations of inaccura- period for which cost data are available, new projects is highly inaccurate and cies, one would expect an improve- accuracy in cost forecasts has not biased, making projects highly risky. ment in accuracy over time, because in improved. Average inaccuracy for rail Comparative studies show that trans- a professional setting errors and their passenger forecasts is –51.4%, with 84% portation projects are no worse than other sources would be recognized and of all rail projects being wrong by more project types in this respect (Flyvbjerg, addressed, for instance, through refer- than ±20%. For roads, average inaccura- Bruzelius, & Rothengatter, 2003). ee processes with scholarly journals cy in traffic forecasts is 9.5%, with half and similar expert critical reviews. of all road forecasts being wrong by Explaining Inaccuracy Undoubtedly, substantial resources more than ±20% (see Table 2). For the Flyvbjerg, Holm, and Buhl (2002; have been spent over several decades 30-year period for which demand data 2004; 2005) and Flyvbjerg and Cowi on improving data and forecasting are available, accuracy in rail and road (2004) tested technical, psychological, models. Nevertheless, this has had no traffic forecasts has not improved. and political-economic explanations effect on the accuracy of forecasts, as When cost and demand forecasts for inaccuracy in forecasting. Technical demonstrated. This indicates that are combined, for instance in the cost- explanations are common in the liter- something other than poor data and benefit analyses that are typically used ature, and they explain inaccuracy in models is at play in generating inaccu- to justify large transportation infrastruc- terms of unreliable or outdated data rate forecasts, a finding that has been corroborated by interviews with fore- Type of Average Standard Level of casters (Flyvbjerg & Cowi, 2004; Project Inaccuracy Deviation Significance Flyvbjerg & Lovallo, in progress; (%) p Wachs, 1990). Psychological and political explana- Rail 44.7 38.4 <0.001 tions better account for inaccurate fore- Bridges and tunnels 33.8 62.4 0.004 casts. Psychological explanations Road 20.4 29.9 <0.001 account for inaccuracy in terms of optimism bias; that is, a cognitive predispo- Source: Flyvbjerg database on large-scale infrastructure projects. sition found with most people to judge Table 1: Inaccuracy in cost forecasts for rail, bridges, tunnels, and roads, respectively future events in a more positive light (construction costs, constant prices) than is warranted by actual experience. Political explanations, on the other Rail Road hand, explain inaccuracy in terms of strategic misrepresentation. Here, when Average inaccuracy (%) -51.4 (sd=28.1) 9.5 (sd=44.3) forecasting the outcomes of projects, Percentage of projects with inaccuracies 84 50 forecasters and managers deliberately larger than ±20% and strategically overestimate benefits Percentage of projects with inaccuracies 72 25 and underestimate costs in order to larger than ±40% increase the likelihood that it is their projects, and not the competition’s, that Percentage of projects with inaccuracies 40 13 larger than ±60% gain approval and funding. Strategic misrepresentation can be traced to polit- Source: Flyvbjerg database on large-scale infrastructure projects. ical and organizational pressures; for Table 2: Inaccuracy in forecasts of rail passenger and road vehicle traffic instance, competition for scarce funds or

6 AUGUST 2006 PROJECT MANAGEMENT JOURNAL jockeying for position. Optimism bias and strategic misrepresentation both Deception involve deception, but where the latter is intentional—i.e., lying—the first is not. Optimism bias is self-deception. Although the two types of explanation are different, the result is the same: inaccurate forecasts and inflated benefit-cost ratios. However, the cures for optimism bias are different from the cures for strategic misrepresentation, as we will see next. Explanatory Power Explanations of inaccuracy in terms of optimism bias have been developed Delusion by Kahneman and Tversky (1979a) and Lovallo and Kahneman (2003). Political and Organizational Pressure Explanations in terms of strategic misrepresentation have been set forth by Figure 1: Explanatory power of optimism bias and strategic misrepresentation, respectively, Wachs (1989; 1990) and Flyvbjerg, in accounting for forecasting inaccuracy as function of political and organizational pressure Holm, and Buhl (2002; 2005). As illustrated schematically in Figure 1, expla- The Planning Fallacy and Reference ation. Human judgment, including nations in terms of optimism bias have Class Forecasting forecasts, is biased. Reference class their relative merit in situations where The theoretical and methodological forecasting is a method for unbiasing political and organizational pressures foundations of reference class forecast- forecasts. are absent or low, whereas such explana- ing were first described by Kahneman Kahneman and Tversky (1979a; tions hold less power in situations and Tversky (1979b) and later by 1979b) found human judgment to be where political pressures are high. Lovallo and Kahneman (2003). generally optimistic due to overconfi- Conversely, explanations in terms of Reference class forecasting was origi- dence and insufficient regard to distri- strategic misrepresentation have their nally developed to compensate for the butional information. Thus, people relative merit where political and orga- type of cognitive bias that Kahneman will underestimate the costs, comple- nizational pressures are high, while they and Tversky found in their work on tion times, and risks of planned become immaterial when such pres- decision-making under uncertainty, actions, whereas they will overestimate sures are not present. Thus the two types which won Kahneman the 2002 Nobel the benefits of the same actions. of explanation complement, rather than prize in economics (Kahneman, 1994; Lovallo and Kahneman (2003, p. 58) compete with one another: one is strong Kahneman & Tversky, 1979a). This call such common behavior the “plan- where the other is weak, and both expla- work showed that errors of judgment ning fallacy” and argue that it stems nations are necessary to understand the are often systematic and predictable from actors taking an “inside view,” phenomenon at hand—the pervasive- rather than random, manifesting bias focusing on the constituents of the ness of inaccuracy in forecasting—and rather than confusion, and that any specific planned action rather than on how to curb it. corrective prescription should reflect the outcomes of similar already-com- In what follows, we present a fore- this. They also found that many errors pleted actions. Kahneman and Tversky casting method called “reference class of judgment are shared by experts and (1979b) argued that the prevalent ten- forecasting,” which bypasses human laypeople alike. Finally, they found dency to underweigh or ignore distrib- bias—including optimism bias and that errors remain compelling even utional information is perhaps the strategic misrepresentation—by cutting when one is fully aware of their nature. major source of error in forecasting. directly to outcomes. In experimental Thus, awareness of a perceptual or cog- “The analysts should therefore make research carried out by Daniel nitive illusion does not by itself pro- every effort to frame the forecasting Kahneman and others, this method has duce a more accurate perception of problem so as to facilitate utilizing all been demonstrated to be more accurate reality, according to Kahneman and the distributional information that is than conventional forecasting methods Tversky (1979b, p. 314). Awareness available,” say Kahneman and Tversky (Kahneman, 1994; Kahneman & may, however, enable one to identify (1979b, p. 316). This may be consid- Tversky, 1979a; 1979b; Lovallo & situations in which the normal faith in ered the single most important piece of Kahneman, 2003). First, we explain the one’s impressions must be suspended advice regarding how to increase accu- theoretical and methodological founda- and in which judgment should be con- racy in forecasting through improved tions for reference class forecasting, then trolled by a more critical evaluation of methods. Using such distributional we present the first instance of reference the evidence. Reference class forecast- information from other ventures simi- class forecasting in project management. ing is a method for such critical evalu- lar to that being forecasted is called

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 7 taking an “outside view,” and it is the months needed to finish and report ones, were overly optimistic. The out- cure to the planning fallacy. Reference the project. The estimates ranged from side view is the one provoked by the class forecasting is a method for sys- 18 to 30 months. One of the team question to the curriculum expert. It tematically taking an outside view on members—a distinguished expert in completely ignored the details of the planned actions. curriculum development—was then project at hand, and it involved no More specifically, reference class posed a challenge by another team attempt at forecasting the events that forecasting for a particular project member to recall as many projects sim- would influence the project’s future requires the following three steps: ilar to theirs as possible, and to think course. Instead, it examined the experi- 1. Identification of a relevant refer- of these projects as they were in a stage ences of a class of similar projects, laid ence class of past, similar proj- comparable to their project. “How out a rough distribution of outcomes ects. The class must be broad long did it take them at that point to for this reference class, and then posi- enough to be statistically mean- reach completion?,” the expert was tioned the current project in that dis- ingful, but narrow enough to be asked. After a while he answered, with tribution. The resulting forecast, as it truly comparable with the spe- some discomfort, that not all the com- turned out, was much more accurate. cific project. parable teams he could think of ever The contrast between inside and 2. Establishing a probability distri- did complete their task. About 40% of outside views has been confirmed by bution for the selected reference them eventually gave up. Of those systematic research (Gilovich, Griffin, class. This requires access to remaining, the expert could not think & Kahneman, 2002). The research credible, empirical data for a of any that completed their task in less shows that when people are asked sim- sufficient number of projects than seven years, nor of any that took ple questions requiring them to take within the reference class to more than 10. The expert was then an outside view, their forecasts become make statistically meaningful asked if he had reason to believe that significantly more accurate. For exam- conclusions. the present team was more skilled in ple, a group of students enrolling at a 3. Comparing the specific project curriculum development than the ear- college were asked to rate their future with the reference class distribu- lier ones had been. The expert said no, academic performance relative to their tion, in order to establish the he did not see any relevant factor that peers in their major. On average, these most likely outcome for the spe- distinguished this team favorably from students expected to perform better cific project. the teams that he had been thinking than 84% of their peers, which is logi- about. His impression was that the cally impossible. The forecasts were Thus, reference class forecasting present team was slightly below aver- biased by overconfidence. Another does not try to forecast the specific age in terms of resources and potential. group of incoming students from the uncertain events that will affect the According to Kahneman, the wise deci- same major were asked about their particular project, but instead places sion at this point would probably have entrance scores and their peers’ scores the project in a statistical distribution been for the team to break up. Instead, before being asked about their expect- of outcomes from the class of reference the members ignored the pessimistic ed performance. This simple diversion projects. In statisticians’ vernacular, information and proceeded with the into relevant outside-view informa- reference class forecasting consists of project. They finally completed the tion, which both groups of subjects regressing forecasters’ best guesses project eight years later, and their were aware of, reduced the second toward the average of the reference efforts went largely wasted—the result- group’s average expected performance class and expanding their estimate of ing curriculum was rarely used. ratings by 20%. That is still overconfi- credible interval toward the correspon- In this example, the curriculum dent, but it is much more realistic than ding interval for the class (Kahneman expert made two forecasts for the same the forecast made by the first group & Tversky, 1979b, p. 326). problem and arrived at very different (Lovallo & Kahneman, 2003, p. 61). Daniel Kahneman relates the fol- answers. The first forecast was the However, most individuals and lowing story about curriculum plan- inside view; the second was the outside organizations are inclined to adopt the ning to illustrate how reference class view, or the reference class forecast. inside view in planning new projects. forecasting works (Lovallo & The inside view is the one that the This is the conventional and intuitive Kahneman, 2003, p. 61). Some years expert and the other team members approach. The traditional way to think ago, Kahneman was involved in a proj- adopted. They made forecasts by focus- about a complex project is to focus on ect to develop a curriculum for a new ing tightly on the project at hand, and the project itself and its details, to subject area for high schools in Israel. considering its objective, the resources bring to bear what one knows about it, The project was carried out by a team they brought to it, and the obstacles to paying special attention to its unique of academics and teachers. In time, the its completion. They constructed in or unusual features, trying to predict team began to discuss how long the their minds scenarios of their coming the events that will influence its future. project would take to complete. progress and extrapolated current The thought of going out and gather- Everyone on the team was asked to trends into the future. The resulting ing simple statistics about related proj- write on a slip of paper the number of forecasts, even the most conservative ects seldom enters a manager’s mind.

8 AUGUST 2006 PROJECT MANAGEMENT JOURNAL This is the case in general, according to First Instance of Reference Class benefits, and duration of large public Lovallo and Kahneman (2003, pp. Forecasting in Practice procurement (HM Treasury, 2003a; 61–62). And it is certainly the case for The first instance of reference class fore- 2003b). cost and demand forecasting in trans- casting in practice may be found in In response to the Treasury’s portation infrastructure projects. Of Flyvbjerg and Cowi (2004): Green Book and its recommendations, the several-hundred forecasts reviewed “Procedures for Dealing with Optimism the U.K. Department for Transport in Flyvbjerg, Bruzelius, and Bias in Transport Planning.”3 Based on decided to collect the type of data Rothengatter (2003) and Flyvbjerg, this study in the summer of 2004, the which the Treasury recommended, and Holm, and Buhl (2002; 2005), not one U.K. Department for Transport and HM on that basis to develop a methodolo- was a reference class forecast.2 Treasury decided to employ the method gy for dealing with optimism bias in Although understandable, project as part of project appraisal for large the planning and management of managers’ preference for the inside view transportation projects. transportation projects. The over the outside view is unfortunate. The immediate background to Department for Transport appointed When both forecasting methods are this decision was the revision to The Bent Flyvbjerg in association with applied with equal skill, the outside Green Book by HM Treasury in 2003 Cowi to undertake this assignment as view is much more likely to produce a that identified for large public pro- regards costing of large transportation realistic estimate. That is because it curement a demonstrated, systematic procurement. The main aims of the bypasses cognitive and political biases tendency for project appraisers to be assignment were two; first, to provide such as optimism bias and strategic overly optimistic: empirically based optimism bias misrepresentation, and cuts directly to “There is a demonstrated, system- uplifts for selected reference classes of outcomes. In the outside view, project atic tendency for project appraisers to transportation infrastructure projects, managers and forecasters are not be overly optimistic. To redress this and, second, to provide guidance on required to make scenarios, imagine tendency, appraisers should make using the established uplifts to pro- events, or gauge their own and others’ explicit, empirically based adjustments duce more realistic forecasts of capital levels of ability and control, so they can- to the estimates of a project’s costs, expenditures in individual projects not get all these things wrong. Human benefits, and duration … [I]t is recom- (Flyvbjerg & Cowi, 2004). Uplifts bias is bypassed. Surely the outside mended that these adjustments be would be established for capital expen- view, being based on historical prece- based on data from past projects or ditures based on the full business case dent, may fail to predict extreme out- similar projects elsewhere” (HM (time of decision to build). comes; that is, those that lie outside all Treasury, 2003b, p. 1). The types of transportation historical precedents. But for most proj- Such optimism was seen as an schemes under the direct and indirect ects, the outside view will produce more impediment to prudent fiscal plan- responsibility of the U.K. Department accurate results. In contrast, a focus on ning, for the government as a whole for Transport were divided into a num- inside details is the road to inaccuracy. and for individual departments within ber of distinct categories in which sta- The comparative advantage of the government. To redress this tendency, tistical tests, benchmarkings, and other outside view is most pronounced for HM Treasury recommended that analyses showed that the risk of cost non-routine projects, understood as appraisers involved in large public pro- overruns within each category may be projects that managers and decision- curement should make explicit, empir- treated as statistically similar. For each makers in a certain locale or organiza- ically based adjustments to the category, a reference class of projects tion have never attempted before—like estimates of a project’s costs, benefits, was then established as the basis for building new plants or infrastructure, or and duration. HM Treasury recom- reference class forecasting, as required catering to new types of demand. It is in mended that these adjustments be by step 1 in the three-step procedure the planning of such new efforts that based on data from past projects or for reference class forecasting previous- the biases toward optimism and strate- similar projects elsewhere, and adjust- ly described. The specific categories gic misrepresentation are likely to be ed for the unique characteristics of the and the types of project allocated to largest. To be sure, choosing the right project at hand. In the absence of a each category are shown in Table 3. reference class of comparative past proj- more specific evidence base, HM For each category of projects, a ref- ects becomes more difficult when man- Treasury encouraged government erence class of completed, comparable agers are forecasting initiatives for departments to collect valid and reli- transportation infrastructure projects which precedents are not easily found; able data to inform future estimates of was used to establish probability distri- for instance, the introduction of new optimism, and in the meantime to use butions for cost overruns for new proj- and unfamiliar technologies. However, the best available data. The Treasury let ects similar in scope and risks to the most projects are both non-routine it be understood that in the future the projects in the reference class, as locally and use well-known technolo- allocation of funds for large public required by step 2 in reference class gies. Such projects are, therefore, partic- procurement would be dependent on forecasting. For roads, for example, a ularly likely to benefit from the outside valid adjustments of optimism in class of 172 completed and compara- view and reference class forecasting. order to secure valid estimates of costs, ble projects was used to establish the

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 9 Category Types of Projects cost overrun, the higher the uplift. For instance, with a willingness to accept a Roads Motorway 50% risk for cost overrun in a road Trunk roads project, the required uplift for this Local roads Bicycle facilities project would be 15%. If the Pedestrian facilities Department for Transport were willing Park and ride to accept only a 10% risk for cost over- Bus lane schemes run, then the required uplift would be Guided buses on wheels 45%. In comparison, for rail, with a Rail Metro willingness to accept a 50% risk for Light rail cost overrun, the required uplift would Guided buses on tracks be 40%. If the Department for Conventional rail High speed rail Transport were willing to accept only a 10% risk for cost overrun, then the Fixed links Bridges required uplift would be 68% for rail. Tunnels All three figures share the same basic S- Building projects Stations shape, but at different levels, demon- Terminal buildings strating that the required uplifts are IT projects IT system development significantly different for different Standard civil engineering Included for reference purposes only project categories for a given level of risk of cost overrun. The figures also Non-standard civil engineering Included for reference purposes only show that the cost for additional reductions in the risk of cost overrun is Table 3: Categories and types of projects used as basis for reference class forecasting different for the three types of projects, with risk reduction becoming increasingly expensive (rising marginal costs) Distribution of Cost Overruns for roads and fixed links below 20% Roads risk, whereas for rail the cost of 100% increased risk reduction rises more 90% slowly, albeit from a high level. 80% Table 4 presents an overview of 70% applicable optimism bias uplifts for 60% the 50% and 80% percentiles for all the project categories listed in Table 3. 50% The 50% percentile is pertinent to the 40% investor with a large project portfolio, 30% Max. Cost Overrun Max. where cost overruns on one project 20% may be offset by cost savings on anoth- Share of Projects with Given 10% er. The 80% percentile—correspon- 0% ding to a risk of cost overrun of -20% 0% 20% 40% 60% 80% 20%—is the level of risk that the U.K. Cost Overrun VS Budget Department for Transport is typically willing to accept for large investments Source: Flyvbjerg database on large-scale infrastructure projects. in local transportation infrastructure. Figure 2: Probability distribution of cost overrun for roads, constant prices (N=172) The established uplifts for optimism bias should be applied to esti- probability distribution of cost overrun is substantial for all three project mated budgets at the time of decision runs shown in Figure 2. The share of types, but highest for rail, followed by to build a project. In the U.K., the projects with a given maximum cost bridges and tunnels, and with the low- approval stage for a large transporta- overrun is shown in the figure. For est risk for roads. tion project is equivalent to the time of instance, 40% of projects have a maxi- Based on the probability distribu- presenting the business case for the mum cost overrun of 10%; 80% of tions described, the required uplifts project to the Department for projects a maximum overrun of 32%, needed to carry out step 3 in a refer- Transport with a view to obtaining the etc. For rail, the probability distribu- ence class forecast may be calculated as go or no-go for that project. tion is shown in Figure 3, and for shown in Figures 5–7. The uplifts refer If, for instance, a group of project bridges and tunnels in Figure 4. The to cost overrun calculated in constant managers were preparing the business figures show that the risk of cost over- prices. The lower the acceptable risk for case for a new motorway, and if they or

10 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Applicable Optimism colleagues were for the projects in the Bias Uplifts reference class would the managers be Category Types of Projects justified in using lower uplifts than 50% 80% those previously described. percentile percentile Conversely, if there is evidence that the Roads Motorway project managers are worse at estimat- Trunk roads ing costs than their colleagues, then Local roads higher uplifts should be used. Bicycle facilities 15% 32% The methodology previously Pedestrian facilities Park and ride described for systematic, practical ref- Bus lane schemes erence class forecasting for transporta- Guided buses on wheels tion projects was developed in Rail Metro 2003–2004, with publication by the Light rail Department of Transport in August Guided buses on tracks 40% 57% 2004. From this date on, local author- Conventional rail ities applying for funding for trans- High speed rail portation projects with the Fixed links Bridges 23% 55% Department for Transport or with HM Tunnels Treasury were required to take into Building projects Stations account optimism bias by using uplifts 4-51%* Terminal buildings as previously described and as laid out IT projects IT system development 10-200%* in more detail in guidelines from the two ministries. Standard civil Included for reference purposes only engineering 3-44%* Forecasting Costs for the Non-standard Included for reference purposes only Edinburgh Tram civil engineering 6-66%* In October 2004, the first instance of *Based on Mott MacDonald (2002, p. 32) no probability distribution available. practical use of the uplifts was record- Table 4: Applicable capital expenditure optimism bia uplifts for 50% and 80% percentiles, ed, in the planning of the Edinburgh constant prices Tram Line 2. Ove Arup and Partners Scotland (2004) had been appointed their client had decided that the risk of budget, then the final budget would be by the Scottish Parliament’s Edinburgh cost overrun must be less than 20%, £420 million. Tram Bill Committee to provide a then they would use an uplift of 32% It follows that the 50% percentile review of the Edinburgh Tram Line 2 on their estimated capital expenditure should be used only in instances business case developed on behalf of budget. Thus, if the initially estimated where investors are willing to take a Transport Initiatives Edinburgh. budget were £100 million, then the high degree of risk that cost overrun Transport Initiatives Edinburgh is the final budget—taking into account will occur and/or in situations where project promoter and is a private limit- optimism bias at the 80%-level— investors are funding a large number ed company owned by the City of would be £132 million (£1 = $1.8). If of projects, and where cost savings Edinburgh Council established to the project managers or their client (underruns) on one project may be deliver major transport projects for the decided instead that a 50% risk of cost used to cover the costs of overruns on Council. The Scottish Executive is a overrun was acceptable, then the uplift other projects. The upper percentiles main funder of the Edinburgh Tram, would be 15% and the final budget (80–90%) should be used when having made an Executive Grant of £115 million. investors want a high degree of certain- £375 million (US$670 million) toward Similarly, if a group of project ty that cost overrun will not occur; for lines 1 and 2, of which Transport managers were preparing the business instance, in stand-alone projects with Initiatives Edinburgh proposed spend- case for a metro rail project, and if they no access to additional funds beyond ing £165 million toward Line 2. or their client had decided that with the approved budget. Other percentiles As part of their review, Ove Arup 80% certainty they wanted to stay may be employed to reflect other assessed whether the business case for within budget, then they would use an degrees of willingness to accept risk Tram Line 2 had adequately taken into uplift on capital costs of 57%. An ini- and the associated uplifts as shown in account optimism bias as regards capi- tial capital expenditure budget of £300 Figures 5–7. tal costs. The business case had esti- million would then become a final Only if project managers have evi- mated a base cost of £255 million and budget of £504 million. If the project dence to substantiate that they would an additional allowance for contin- managers or their client required only be significantly better at estimating gency and optimism bias of £64 mil- 50% certainty they would stay within costs for the project at hand than their lion—or 25%—resulting in total

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 11 distributional information made avail- Distribution of Cost Overruns Rail able by the U.K. Department for 100% Transport, Ove Arup was able to take 90% an outside view on the Edinburgh Tram Line 2 capital cost forecast and 80% thus unbias what appeared to be a 70% biased forecast. As a result, Ove Arup’s 60% client, The Scottish Parliament, was 50% provided with a more reliable estimate 40% of what the true costs of Line 2 was 30% likely to be. Max. Cost Overrun Max. 20%

Share of Projects with Given Potentials and Barriers for Reference 10% Class Forecasting 0% As previously mentioned, two types of -20% 0% 20% 40% 60% 80% 100% explanation best account for forecasting Cost Overrun VS Budget inaccuracy: optimism bias and strategic misrepresentation. Reference class fore- Source: Flyvbjerg database on large-scale infrastructure projects. casting was originally developed to miti- Figure 3: Probability distribution of cost overrun for rail, constant prices (N=46) gate optimism bias, but reference class forecasting may help mitigate any type of capital costs of approximately £320 mism bias uplifts be applied to the human bias, including strategic bias, million. Ove Arup concluded about budget at the time of decision to build, because the method bypasses such bias this overall estimate of capital costs which typically equates to business by cutting directly to empirical outcomes that it seemed to have been rigorously case submission. (In addition, Tram and building forecasts on these. Even so, prepared using a database of costs, Line 2 had not yet even reached the the potentials for and barriers to refer- comparison to other U.K. light rail outline business case stage, indicating ence class forecasting will be different in schemes, and reconciliations with ear- that risks would be substantially high- situations in which (1) optimism bias is lier project estimates. Ove Arup found, er at this early stage, as would corre- the main cause of inaccuracy as com- however, that the following potential sponding uplifts. On that basis, Arup pared to situations in which (2) strategic additional costs needed to be consid- concluded that “it is considered that misrepresentation is the reason for inac- ered in determining the overall capital current optimism bias uplifts [for Tram curacy. We therefore need to distinguish costs: £26 million for future expendi- Line 2] may have been underestimat- between these two types of situations ture on replacement and renewals and ed” [Ove Arup & Partners Scotland, when endeavoring to apply reference £20 million as a notional allowance 2004, p. 27].) class forecasting in practice. for a capital sum to cover risks of Finally, Ove Arup mentioned that In the first type of situation—in future revenue shortfalls, amounting the Department for Transport guid- which optimism bias is the main cause to an increase in total capital costs of ance does allow for optimism bias to of inaccuracy—we may assume that 14.4% (Ove Arup and Partners be adjusted downward if strong evi- managers and forecasters are making Scotland, 2004, pp. 15–16). dence of improved risk mitigation can honest mistakes and have an interest Using the U.K. Department for be demonstrated. According to Ove in improving accuracy. Consider, for Transport uplifts for optimism bias Arup, this may be the case if advanced example, the students who were asked previously presented on the base costs, risk analysis has been applied, but this to estimate their future academic per- Ove Arup then calculated the 80th per- was not the case for Tram Line 2. Ove formance relative to their peers. We centile value for total capital costs— Arup therefore concluded that “the may reasonably believe that the stu- the value at which the likelihood of justification for reduced Department dents did not deliberately misrepresent staying within budget is 80%—to be for Transport optimism bias uplifts their estimates, because they had no £400 million (i.e., £255 million x would appear to be weak” (Ove Arup interest in doing so and were not 1.57). The 50th percentile for total & Partners Scotland, 2004, pp. exposed to pressures that would push capital costs—the value at which the 27–28). Thus, the overall conclusion them in that direction. The students likelihood of staying within budget is of Ove Arup was that the promoter’s made honest mistakes, which pro- 50%—was £357 million (i.e., £255 capital cost estimate of approximately duced honest, if biased, numbers million x 1.4). Ove Arup remarked that £320 million was optimistic. Most regarding performance. And, indeed, these estimates of total capital costs likely Tram Line 2 would cost signifi- when students were asked to take into were likely to be conservative—that is, cantly more. account outside-view information, we low—because the U.K. Department for By framing the forecasting prob- saw that the accuracy of their estimates Transport recommends that its opti- lem to allow the use of the empirical improved substantially. In this type of

12 AUGUST 2006 PROJECT MANAGEMENT JOURNAL situation—when forecasters are hon- Distribution of Cost Overruns estly trying to gauge the future—the Fixed Links potential for using the outside view 100% and reference class forecasting will be 90% good. Forecasters will be welcoming 80% the method and barriers will be low, 70% because no one has reason to be against a methodology that will 60% improve their forecasts. 50% In the second type of situation— 40% in which strategic misrepresentation is 30% the main cause of inaccuracy—differ- Cost Overrun Max. 20% ences between estimated and actual Share of Projects with Given 10% costs and benefits are best explained by political and organizational pres- 0% -20% 30% 80% 130% 180% sures. Here, managers and forecasters would still need reference class fore- Cost Overrun VS Budget casting if accuracy were to be Source: Flyvbjerg database on large-scale infrastructure projects. improved, but managers and forecast- Figure 4: Probability distribution of cost overrun for fixed links, constant prices (N=34) ers may not be interested in this because inaccuracy is deliberate. Biased forecasts serve strategic purpos- Required Uplift es that dominate the commitment to Roads accuracy and truth. Consider, for 70% example, city managers with responsi- 60% bility for estimating costs and benefits of urban rail projects. Here, the 50% assumption of innocence regarding outcomes typically cannot be upheld. 40% Cities compete fiercely for approval 30% and for scarce national funds for such projects, and pressures are strong to Required Uplift 20% present projects as favorably as possible; that is, with low costs and high 10% benefits, in order to beat the competi- 0% tion. There is no incentive for the indi- 0% 10% 20% 30% 40% 50% 60% vidual city to unbias its forecasts, but Acceptable Chance of Cost Overrun quite the opposite. Unless all other cities also unbias, the individual city Source: Flyvbjerg database on large-scale infrastructure projects. would lose out in the competition for Figure 5: Required uplift for roads as function of the maximum acceptable level of risk for cost funds. Project managers are on record overrun, constant prices (N=172) confirming that this is a common situation (Flyvbjerg & Cowi, 2004, pp. reward accurate forecasts and punish higher the stakes, and the higher the 36–58; Flyvbjerg & Lovallo, in inaccurate ones. Forecasters and pro- level of political and organizational progress). The result is the same as in moters should be made to carry the pressures, the more pronounced the case of optimism: actors promote full risks of their forecasts. Their will be the need for such measures ventures that are unlikely to perform as work should be reviewed by inde- of accountability. Flyvbjerg, promised. But the causes are different, pendent bodies such as national Bruzelius, and Rothengatter (2003) as are possible cures. auditors or independent analysts, and Flyvbjerg, Holm, and Buhl In this type of situation, the and such bodies would need refer- (2005) further detailed the design potential for reference class forecasting ence class forecasting to do their of such measures and how they may is low—the demand for accuracy is work. Projects with inflated bene- be implemented in practical project simply not there—and barriers are fit-cost ratios should be stopped or management. high. In order to lower barriers, and placed on hold. Professional and The existence of strategic misrepre- thus create room for reference class even criminal penalties should be con- sentation does not exclude the simul- forecasting, measures of accountability sidered for people who consistently taneous existence of optimism bias, must be implemented that would produce misleading forecasts. The and vice versa. In fact, it is realistic to

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 13 tion is perceived as being available to Required Uplift Rail the project and its contractors. This 90% makes it important to combine the introduction of reference class fore- 80% casting and optimism bias uplifts with 70% tight contracts, and maintained incen- 60% tives for promoters to undertake good 50% quantified risk assessment and exercise prudent cost control during project 40% implementation. How this may be

Required Uplift 30% done is described in Flyvbjerg and 20% Cowi (2004). 10% Notes 0% 1 Inaccuracy is measured in percent as 0% 10% 20% 30% 40% 50% 60% (actual outcome/forecast outcome -1) x Acceptable Chance of Cost Overrun 100. The base year of a forecast for a

Source: Flyvbjerg database on large-scale infrastructure projects. project is the time of decision to build that project. An inaccuracy of 0 indi- Figure 6: Required uplift for rail as function of the maximum acceptable level of risk for cost cates perfect accuracy. Cost is measured overrun, constant prices (N=46) as construction costs. Demand is measured as number of vehicles for roads Required Uplift and number of passengers for rail. Fixed Links 160% 2 The closest thing to an outside view 140% in large infrastructure forecasting is Gordon and Wilson’s (1984) use of 120% regression analysis on an international 100% cross section of light-rail projects to 80% forecast patronage in a number of light-rail schemes in North America. 60%

Required Uplift 3 40% The fact that this is, indeed, the first instance of practical reference class 20% forecasting has been confirmed with 0% Daniel Kahneman and Dan Lovallo, 0% 10% 20% 30% 40% 50% 60% who also knows of no other instances Acceptable Chance of Cost Overrun of practical reference class forecasting. Personal communications with Daniel Source: Flyvbjerg database on large-scale infrastructure projects. Kahneman and Dan Lovallo, author’s

Figure 7: Required uplift for fixed links as function of the maximum acceptable level of risk for archives. cost overrun, constant prices (N=34) References expect such co-existence in forecasting ciencies and overspending. Reserves Altshuler, A., & Luberoff, D. in large and complex projects and will be spent simply because they are (2003). Mega-projects: The changing poli- organizations. This again underscores there, as the saying goes in the con- tics of urban public investment. the point that improved forecasting struction business. For instance, it is Washington, DC: Brookings Institution. methods—here, reference class fore- important to recognize that, for the American Planning Association. casting—and measures of accountabil- previously mentioned examples, the (2005). JAPA article calls on planners to ity must go hand in hand if the introduction of reference class fore- help end inaccuracies in public project rev- attempt to arrive at more accurate forecasting and optimism-bias uplifts enue forecasting. Retrieved April 7, 2005, casts is to be effective. would establish total budget reserva- from http://www.planning.org/newsre- Finally, it could be argued that in tions (including uplifts) which for leases/2005/ftp040705.htm some cases the use of reference class some projects would be more than Flyvbjerg, B. (2005, Spring/Summer). forecasting may result in such large adequate. This may in itself create an Design by deception: The politics of reserves set aside for a project that this incentive which works against firm megaproject approval. Harvard Design would in itself lead to risks of ineffi- cost control if the total budget reserva- Magazine, 22, 50–59.

14 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Flyvbjerg, B., Bruzelius, N., & Gilovich, T., Griffin, D., & corrective procedures. In S. Makridakis & Rothengatter, W. (2003). Megaprojects and Kahneman, D. (Eds.). (2002). S. C. Wheelwright (Eds.), Studies in the risk: An anatomy of ambition. Cambridge, Heuristics and biases: The psychology of management sciences: Forecasting (p. 12) MA: Cambridge University Press. intuitive judgment. Cambridge, MA: Amsterdam: North Holland. Flyvbjerg, B., & Cowi. (2004). Cambridge University Press. Lovallo, D., & Kahneman, D. Procedures for dealing with optimism bias Gordon, P., & Wilson, R. (1984). (2003, July). Delusions of success: in transport planning: Guidance docu- The determinants of light-rail transit How optimism undermines executives’ ment. London: UK Department for demand: An international cross-sec- decisions. Harvard Business Review, Transport. tional comparison. Transportation 56–63. Flyvbjerg, B., Holm, M. K. S., & Research A, 18A(2), 135–140. MacDonald, M. (2002). Review of Buhl, S. L. (2002, Summer). HM Treasury. (2003a). The green large public procurement in the UK. Underestimating costs in public works book: Appraisal and evaluation in central gov- Study for HM Treasury. London: HM projects: Error or lie? Journal of the ernment, treasury guidance. London: TSO. Treasury. American Planning Association, 68(3), HM Treasury. (2003b). Ove Arup & Partners Scotland. 279–295. Supplementary green book guidance: (2004). Scottish parliament, Edinburgh Flyvbjerg, B., Holm, M. K. S., & Optimism bias. London: HM Treasury. tram line 2 review of business case. West Buhl, S. L. (2004). What causes cost Kahneman, D. (1994). New chal- Lothian, Scotland: Author. overrun in transport infrastructure lenges to the rationality assumption. Vanston, J. H., & Vanston, L. K. projects? Transport Reviews, 24(1), Journal of Institutional and Theoretical (2004). Testing the tea leaves: 3–18. Economics, 150, 18–36. Evaluating the validity of forecasts. Flyvbjerg, B., Holm, M. K. S., & Kahneman, D., & Lovallo, D. Research-Technology Management, Buhl, S. L. (2005, Spring).How (1993). Timid choices and bold fore- 47(5), 33–39. (in)accurate are demand forecasts in casts: A cognitive perspective on risk Wachs, M. (1989). When planners public works projects? The case of taking. Management Science, 39, 17–31. lie with numbers. Journal of the transportation. Journal of the American Kahneman, D., & Tversky, A. American Planning Association, 55(4), Planning Association, 71(2), 131-146. (1979a). Prospect theory: An analysis 476–479. Flyvbjerg, B., & Lovallo, D. (n.d.). of decisions under risk. Econometrica, Wachs, M. (1990). Ethics and Delusion and deception: Two models for 47, 313–327. advocacy in forecasting for public pol- explaining executive disaster. Kahneman, D., & Tversky, A. icy. Business and Professional Ethics Unpublished manuscript. (1979b). Intuitive prediction: Biases and Journal, 9(1–2), 141–157.

BENT FLYVBJERG is professor of planning at the Department of Development and Planning at Aalborg University, Denmark. He has doctor of technology from Aalborg University and holds a PhD in urban geography and planning from Aarhus University, Denmark. He was twice a visiting Fulbright Scholar to the U.S., where he did research at the University of California at Los Angeles and Berkeley and at Harvard University. He has been a visiting fellow with the European University Institute in Florence. He has written several books: Megaprojects and Risk: An Anatomy of Ambition, Making Social Science Matter, and Rationality and Power: Democracy in Practice. His books and articles have been translated into 15 languages. His main research interest is urban policy and planning. He is currently conducting research on megaprojects, phronetic planning research, and the relationship between truth and lying in policy and planning. He has two decades of practical experience from working as a policy and planning adviser and consultant to more than 30 public and private organizations, including the EU Commission, the United Nations, national and local government, auditors general, and private companies. His work covers both developed and developing nations. He has been adviser to the government of Denmark in formulating national policies for transportation, environment, and research. He is founding chairman of the Geography Program at Aalborg University, co-founder of the university’s Program in Planning and Environment, and founding director of the university’s Research Program on Large Infrastructure Projects. He has received numerous honors and awards, including the Danish National Science Council Distinguished Research Scholarship (equivalent to the MacArthur Fellowship). In 2002, Queen Margrethe II of Denmark conferred upon Bent Flyvbjerg the Knighthood of the Order of the Dannebrog for his professional accomplishments.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 15 AN EMPIRICAL INVESTIGATION OF THE SPONSOR’S ROLE IN PROJECT INITIATION

TIMOTHY J. KLOPPENBORG, PMP, Xavier University DEBORAH TESCH, Xavier University CHRIS MANOLIS, Xavier University MARK HEITKAMP, PMP, American Modern Insurance Group

Introduction ABSTRACT lthough most organizational projects have many interested parties or stakeholders, someone must, in all projects, take the primary role of sponsor- Although most organizational projects A ship. The senior executive—the one who “owns” a project and is considered have many interested parties or stakeholders, the executive with the fiscal responsible for ensuring its success—is typically the one who proposes a project authority, political clout, and personal and whose business unit will reap the benefits of a successfully completed project. commitment to see a project through is The effectiveness of this senior executive, often labeled the executive sponsor, is the project sponsor. Certainly, there are frequently a predictor of project success. Involved and committed executive spon- project sponsor tasks associated with sors must have enough clout to make the changes that are deemed necessary to the successful completion of a project. Yet, very little research exists that successfully complete a project (Perkins, 2005). Many of these changes must occur attempts to identify and validate a set of during the initiating stage of the project life cycle. Interestingly, very little research executive sponsor behaviors necessary exists specifying exactly what tasks or behaviors constitute the role of an effective for successful project implementation. executive sponsor and ultimately a successful project. Using a previously established database The basis of this research is to identify and validate empirically critical spon- of project risk avoidance and mitigation strategies that was supplemented by an sor behaviors. In the paper, we describe the mechanisms employed to initially updated literature search, we examine identify effective sponsor behaviors associated with the initiating stage of a proj- and classify behaviors associated with ect. Once identified, these behaviors are subjected to a validation and prioritiza- the role of a project sponsor. The scope tion process utilizing an established procedure. Next, an exploratory factor of this exploratory research includes: (1) analysis of the behaviors is conducted followed by tests of association between the identifying project sponsor-related behaviors; (2) validating and prioritizing project sponsor behaviors and project outcomes. the sponsor behaviors utilizing an estab- The paper is organized in the following manner: The first section outlines the cur- lished procedure; (3) empirically validat- rent status of research on the role of the executive sponsor in projects. Next, the ing the behaviors; and (4) empirically research methodology is presented, followed by the results of empirical analyses. The testing the association of the project paper ends with a discussion of the results and recommendations for project sponsors. sponsor behaviors with various dimensions of project success (project outcomes). Background Sponsor Behavior Keywords: project sponsor; project suc- Although most project management literature recognizes the project sponsor as cess; sponsorship a key stakeholder in most projects, very little research has examined the role of

©2006 by the Project Management Institute the executive sponsor in achieving project success. Project sponsors are primarily Vol. 37, No. 3, 16-25, ISSN 8756-9728/03 considered responsible for project resources (Helm & Remington, 2005). This is evident in the definition of a sponsor as provided in the PMBOK® Guide: “the person or group that provides the financial resources, in cash or in kind, for the project” (Project Management Institute [PMI], 2004, p. 26). Additionally, the

16 AUGUST 2006 PROJECT MANAGEMENT JOURNAL project manager is considered the primary risk taker for the project is formally authorized, often in the form of a whom the project is undertaken (Association for Project signed charter. We focus exclusively on the initiating stage Management, 2005). based on, one, the belief that sponsors have a more direct role Recent anecdotal evidence, as stated in articles such as in the initiating phase of a project (versus later stages), and, “The Elusive Executive Sponsor” (Perkins, 2005), “Surviving two, the importance of getting a project off to a good start. the Sponsor Exit” (Melymuka, 2004a), and “Firing Your Project Sponsor” (Melymuka, 2004b), stress the role of the Dimensions of Project Success project sponsor with respect to project success, often quoting DeLone and McLean (1992) identified key indicators of advice from senior project managers about how to deal with information technology (IT) project success that form an inadequate project sponsors. Helm and Remington (2005) initial framework for measuring system performance. This undertook a combined analysis of the roles and responsibil- framework expanded on the early “triple constraint” notion ities of the project sponsor in relation to the project organi- of time, cost, and performance to consider information and zational structure and the behavior and practices of key system quality dimensions as well as use and satisfaction identified agents. Based on evidence from the literature, constraints. In addition, the impact of the system was exam- questions were formed into guided in-depth interviews with ined from the perspective of its impact on individuals (often selected project personnel. Project managers and sponsors customers) and the organization. were asked to define the role of the project sponsor and the Shenhar, Tishler, Dvir, Lipovetsky, and Lechler (2002) way that that role contributed to project success. The most considered 13 success measures from a previous implemen- frequently cited project sponsor characteristics included: tation grouped into three dimensions: meeting design goals, 1. Appropriate seniority and power in the organization benefits to customers, and commercial success and future 2. Political knowledge and savvy potential. Pinto (2004) included the time-dependent 3. Ability/willingness to make project/organization dimension considered in Shenhar, Tishler, Dvir, Lipovetsky, connections and Lechler (2002) as a separate dimension in assessing the 4. Courage/willingness to go to battle with others on effectiveness of a project, indicating that we must not only behalf of the project evaluate projects in current terms, but also look to the future 5. Ability to motivate the team and provide ad hoc sup- potential that a project offers in terms of generating new port to the team business and new opportunities. The four relevant dimen- 6. Willingness to partner with the project team and sions of success, as described by Pinto (2004), include proj- project manager ect efficiency, impact on the customer, business success, and 7. Excellent communication skills future potential. In the current study, we adopt this contri- 8. Personally compatible with other key players bution to the literature in establishing project success or 9. Ability/willingness to challenge the project and provide outcome measures. objectivity (Helm & Remington, 2005) Research Objectives In the public sector, the project sponsor is described as In this research, we propose to test the effects of rigorously the person responsible for representing the public client and identified project sponsor behaviors on project outcomes. acting as a day-to-day manager of the client’s interests with- That is, based on the previous discussion, we predict that in the project. Hall, Holt, and Purchase (2002) conducted a derived sponsor behaviors will be significantly associated series of interviews considering the role of the project spon- with various project outcomes. sor in areas where New Public Management1 is being practiced. An analysis of the sponsor interviews revealed the Methodology complexity of the sponsor’s role. Unable to consider any Identifying Sponsor Behavior one theme in isolation, sponsors are at once involved with The role of the executive sponsor is best examined in con- juggling multiple needs of stakeholders and user groups, text. The present research started by exploring the role of the departmental procedures, and government edicts while executive sponsor in the context of IT and project risk with- dealing with a legacy of mistrust and adversarial contracts. in specific stages of a project’s life cycle. In order to increase The dominant recommendation for these sponsors was to the success rate of information systems (IS) projects, man- develop a mechanism for dealing with the “softer” cultural agers must identify risks, understand their consequences, and attitudinal issues culminating in a mechanism to and plan to either avoid or mitigate their impact (Royer, encourage dialog and promote cooperation. Thus, in order 2000; Project Management Institute, 2004). Much of the to cope with the variety of demands, project sponsors need identified project risk literature in the IS/IT area has focused to develop long-term relationships with constituents and on the identification and/or quantification of risk factors. acquire significant experience per their role. Once these factors are identified, researchers often attempt For this study, we consider the four stages of initiating, to develop risk management frameworks. From a compre- planning, executing, and closing to describe the project life hensive list of risk factors, a set of nine risk factor groups was cycle. We limit our evaluation to the initiating stage, which developed (Kloppenborg & Tesch, 2004) that are consistent starts when the project idea is first identified and ends when with previous research—particularly the five risk groups pre-

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 17 viously established by Barki, Rivard, and Talbot (1993) and ly the same, resulting in an elimination of one behavior; the 14 groups established by Schmidt, Lyytinen, Keil, and two, the determination of which behaviors had minor dif- Cule (2001). The risk groups used in the second study ferences and thereby should be retained; and, three, ensur- included corporate environment, sponsorship/ownership, ing that wording was clear for all behavior statements. At relationship management, project management, scope, least two experts were required to agree on each behavior. As requirements, funding, scheduling, development of process, a result, 22 behaviors were removed as duplicates, and vari- personnel, staffing, technology, external dependencies, and ous revisions were made to other behavior statements. This planning. The background work just described was our first phase resulted in a list of 72 behaviors. The next logical pro- step toward identifying successful sponsor behavior. gression is to establish and confirm empirically this list of Using the list of (identified, classified, and quantified) initiating sponsor behaviors, and to test their association risk factors, a group of project management professionals with various project outcomes. in a local Project Management Institute (PMI) professional development seminar were asked to define avoidance The Sample and mitigation behaviors associated with identified risks. To establish and confirm empirically the list of 72 initiating A total of 745 behaviors were developed and stored in a sponsor behaviors, we constructed a survey and utilized a database. Each of these behaviors was then applied to one convenience sample of 109 respondents that was developed of seven project leadership task areas in each of the four from contacts at Project Management Institute conferences project life cycle stages (Kloppenborg, Shriberg, & and other professional settings. Demographics for the same Venkaraman, 2003). Furthermore, the leadership role may be found in Figure 1. Approximately 60% of the respon- (sponsor, project manager, etc.) that had responsibility for dents described their field of work as general management, the behavior was identified. This categorization of risk information systems, or operations. Similarly, about 60% avoidance/mitigation strategies in terms of project leader- said their industry was consulting, manufacturing, service, or ship offers project management professionals a framework education. About half said projects in their organization useful in examining successful executive sponsor behav- averaged less than one year in length, and just over 25% held iors. Of these identified behaviors, 142 are associated with a PMP certification. While the level of experience varied steering teams and sponsors specifically during the initiat- greatly, about 29% had 30 or more years of experience. The ing stage (Kloppenborg & Tesch, 2004). In a subsequent respondents are largely from North America. In summary, literature review, we considered all types of projects and the sample is quite diverse in several ways. added eight new initiating stage sponsor behaviors for a total of 150 avoidance/mitigation behaviors. Ultimately, Procedure and Measures the behaviors were described in short written statements Respondents were administered an online survey and asked and arranged in a survey-like format. to consider behaviors that a project sponsor might engage In the next phase of the research, a select group of five sen- in during the project initiation stage in order to facilitate ior project management professionals (PMPs) examined the the success of a project. Project initiation was defined as list of behavior statements in order to eliminate redundancy, beginning with the idea for a potential project and ending validate behaviors, and simplify narrative descriptions. These with a commitment, often in the form of a charter that is professionals, or experts, collectively represent 167 years of signed by both the sponsor and the project team. The spon- business experience, much of which is in project management, sor was described to respondents as a senior executive who and have worked in insurance, health care, information sys- has an interest in the results of a project. This executive may tems, government, military, consulting, and manufacturing also have monetary control over the project. Often, the environments. The PMPs used the method for priority mark- executive has organizational clout but does not have signif- ing (MPM); this is an established technique useful in reducing icant time to personally manage the project. large amounts of language data to a vital subset (Center for Respondents were asked to rate a series of 72 sponsor Quality of Management [CQM], 1997). According to MPM, behavior statements using a Likert-type response scale teams or individuals select the best qualitative data by elimi- ranging from 1 (strongly disagree) to 7 (strongly agree). For nating non-candidates in several phases (CQM, 1997). PMPs each behavior, respondents were asked to consider how were asked to mark each behavior that they considered to be important the behavior is with respect to facilitating a suc- associated with a sponsor and had at least a moderate impact cessful project. on project success. Each expert considered each behavior at Respondents were next asked to consider 13 perform- least twice. Fifty-six behaviors that all five experts agreed upon ance dimensions as described in Pinto (2004) that are that either had little impact on project success, should not have often used to describe the value of a project upon comple- happened during the initiating phase, or that sponsors had no tion, or project outcomes. For each outcome or outcome responsibility for, were deleted. statement, respondents considered the extent to which the Next, the PMPs grouped the behaviors into sets of sim- outcome is significant in terms of a project’s success. The ilar groups for the purpose of further eliminating duplicate 13 outcome items also followed a 7-point, Likert-type behaviors. The difficult part of the deduplicating effort was, response scale ranging from 1 (strongly disagree) to 7 one, the determination of which similar behaviors were real- (strongly agree).

18 AUGUST 2006 PROJECT MANAGEMENT JOURNAL N 109 Sex Males 78 71.6% Experience 1 - 5 years 2 1.8% Females 24 22.0% 6 - 10 years 10 9.2% No response 7 6.4% 11 - 15 years 9 8.3% 16 - 20 years 14 12.8% Level Front line 5 4.6% 21 - 25 years 19 17.4% First level 6 5.5% 26 - 30 years 17 15.6% Second level 18 16.5% > 30 years 32 29.4% Executive 46 42.2% No response 6 5.5% Other 27 24.8% No response 7 6.4% Duration < 4 months 6 5.5% 4 mo to 1 yr 52 47.7% Field ACCT 1 0.9% 1 - 2 years 24 22.0% FIN 4 3.7% 2 - 3 years 11 10.1% GM 27 24.8% 3 - 5 years 9 8.3% HR 5 4.6% 6 or more yrs 0 0.0% IS 20 18.3% No response 7 6.4% MKT 2 1.8% OP 19 17.4% Certification Certified 29 26.6% Other 25 22.9% Pursuing certification 5 4.6% No response 6 5.5% Not certified 69 63.3% No response 6 5.5% Industry Consulting 21 19.3% Construction 2 1.8% Region Eastern Europe 1 0.9% Education 12 11.0% Western Europe 10 9.2% Engineering 7 6.4% North America – 15 13.8% Canada Manufacturing 18 16.5% North America - US 71 65.1% Retail 5 4.6% South America 1 0.9% Service 15 13.8% Asia – China 0 0.0% Other 23 21.1% Asia – India 1 0.9% No response 6 5.5% Asia – other 2 1.8% Africa 2 1.8% Australia/New Zealand 0 0.0% No response 6 5.5%

Figure 1: Sample demographics Questionnaire Data tions and commitment In order to reduce the number of variables on the question- 2. Alignment—Eight items asking respondents about the naire, principle components analyses (with varimax rota- importance of defining and aligning project commitment tion) were conducted separately for the sponsor behavior 3. Performance—Four items asking respondents about the and project outcome statements (items). Based on these importance of a sponsor defining performance/success analyses and a priori reasoning, eight behavior and three standards on behalf of a project manager outcome composite variables or factors were created. The 4 Project manager—Three items asking respondents items constituting each factor are depicted in Figure 2. The about the importance of a project sponsor both factors, some of which are explained in greater detail later in selecting and mentoring project managers the paper, include the following: 5. Prioritize—Five items asking respondents about the importance of prioritizing tasks Sponsor Behaviors Factors 6. Teams—Ten items asking respondents about the 1. Commitment—Eleven items asking respondents importance of selecting and establishing project about the importance of establishing communica- teams

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 19 Factor Items Constituting Factor Cronbach’s Coefficient Alpha

Establishing Communicating support for a project is important 0.90 communications Demonstrating the proper level of commitment to a project is important and commitment Understanding the expectations of management is important Making sure a project has the support of management is important Ensuring that executives are committed to a project is important Personally demonstrating appropriate levels of participation in a project is important Ensuring that identified stakeholders support the project is important Ensuring that plans are communicated with stakeholders is important Ensuring that communication procedures with management are established is important Demonstrating a high enough level of commitment to a project is important

Defining and Aligning the objectives/goals of a project with the objectives/goals of a firm is important 0.86 aligning the Ensuring that expected project benefits to the business are defined is important project Aligning project scope and funding is important Ensuring that project goals and objectives are clearly defined is important Validating project priority in terms of business value is important Ensuring that a project’s goals and success factors are clearly defined is important Personally communicating the strategic value of a project is important Ensuring that the scope of a project is clearly defined is important

Selecting and Staffing a project with people who have appropriate skills is important 0.91 establishing When selecting team members and subject-matter experts for a project, understanding the project teams required skills necessary to ensure project success is important Selecting people with proper knowledge is important Ensuring that a team has proper training and tools is important Ensuring that all parties involved know and understand their personal responsibilities is important Selecting people with appropriate people skills is important Ensuring that project managers quickly resolve issues that could hinder the performance of a project is important Ensuring that regular meetings to review the status of a project are held is important Ensuring that team operating procedures are included in a charter is important Ensuring written documentation of required involvement by all parties is important

Risk planning Ensuring that risks are identified is important 0.79 Ensuring a risk assessment plan is developed is important Ensuring that project risks are analyzed is important

Figure 2: Composite variables: Behavior and outcomes

20 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Factor Items Constituting Factor Cronbach’s Coefficient Alpha

Establishing Ensuring that a formal change process is in place is important 0.78 change control Ensuring that project goals and objectives are approved is important Ensuring that a “change control board” is established is important Ensuring that any and all changes are noted and understood is important Ensuring all parties agree on project scope is important

Defining Ensuring that metrics to measure a project’s success are established is important 0.70 performance/ Ensuring that the strategic value of a project is communicated is important success Empowering project managers so that they can do their job effectively is important Defining a project manager’s performance expectations is important

Prioritizing Ensuring that a “project feasibility study” is conducted that includes expected benefits to 0.77 the business is important It is important that general project goals and objectives are agreed upon prior to more detailed planning Ensuring that a steering committee prioritizes projects is important Ensuring that all stakeholders of a project are identified is important Emphasizing the benefits of the project to the steering team is important

Selecting and Helping the project manager develop people skills is important 0.77 mentoring the Monitoring a project manager’s performance is important project manager Helping the project manager understand the “big picture” is important

Future Increasing market share 0.93 (outcome) Opening new lines of products Opening new markets Generating a large market share Developing a new technology Achieving significant commercial success

Meeting Meeting schedule expectations 0.83 agreements Meeting budget (outcome) Finishing a project on time Meeting technical specifications

Customer Creating a project that leads to enhanced satisfaction on the part of the customer 0.75 (outcome) Creating a project that is used by the client Addressing customer needs

Figure 2 (cont.): Composite variables: Behavior and outcomes

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 21 7. Risk—Three items asking respondents about the ranged from .70 to .93, with a mean of .81. Although not importance of risk planning (i.e., predicting and depicted in Figure 2, each of the sponsor behavior factors assessing risk) were significantly and positively correlated (p’s < .05) 8. Change—Five items asking respondents about the with one another, and all but two of the outcome factors importance of establishing change control (i.e., hav- were significantly and positively correlated (p’s < .05) ing procedures in place for handling change) with one another.

Outcome Factors Results 9. Agreements—Six items asking respondents about the To test the association or relationship between the spon- importance of meeting agreements (budgets, sched- sor behavior and the project outcome variables, correla- uling expectations, etc.) tional analysis was utilized. Illustrated in Figure 3, the 10. Customer—Four items asking respondents about results reveal several significant and positive correla- the importance of pleasing the customer (customer tions. First, the agreements outcome variable was posi- satisfaction) tively and significantly related to the commitment, 11 . Future—Three items asking respondents about the alignment, performance, and project manager behavior importance of creating future benefits (commercial variables. Next, the customer outcome variable was pos- success, increased market share, new products and itively and significantly related to the alignment, per- technologies, etc.). formance, project manager, prioritize, and teams behavior variables. And, finally, the future outcome vari- In order to test the internal-consistency reliability of able was positively and significantly related to the per- these factors, Cronbach’s coefficient alphas were comput- formance, project manager, prioritize, and teams ed (see Figure 2). The reliabilities for these 11 variables behavior variables.

Behaviors Outcomes

Establishing commitment (r = .23, p < .03) and communications Meeting Agreements: specifications, (r = .24, p < .01) schedule

Aligning and defining (r = .28, p < .01) the project (r = .22, p < .03)

(r = .28, p < .01) Defining performance/success (r = .43, p < .00) Customer: needs, (r = .41, p < .00) use Selecting and mentoring the project manager (r = .29, p < .00) (r = .29, p < .00)

(r = .26, p < .01) Prioritizing (r = .38, p < .00) Future: (r = .31, p < .00) commercial success, Selecting and establishing market share, project teams (r = .33, p < .00) new products

Figure 3: Significant associations between sponsor behavior and project outcomes

22 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Discussion project manager. This consists of behaviors such as help- Based upon a very thorough and exhaustive process, this ing the project manager develop people skills and moni- research accurately identified and measured both project toring the project manager’s performance. An effective sponsor behaviors and project outcomes. The identified project manager must be helpful in achieving agreements, constructs demonstrate satisfactory levels of both reliabil- and the sponsor mentoring the project manager is intend- ity and validity. The research also tested associations ed to help. between these behaviors and desired project outcome variables. This analysis provides important insight into Customer Success behaviors that executive sponsors might exhibit if they The outcome entitled the customer includes meeting the wish to attain successful project outcomes. customer’s needs (whether they are specified or not), cre- Each of six behavior factors (establishing communi- ating a project result that the customer uses, and enhanc- cations and commitment, defining and aligning the proj- ing customer satisfaction. The sponsor behaviors ect, defining performance/success, mentoring the project associated with customer success include defining and manager, prioritizing and selecting, and establishing proj- aligning the project, defining project performance and ect teams) is significantly correlated with at least one of success, mentoring the project manager, prioritizing the the three outcome measures (meeting agreements, cus- project, and selecting and establishing the project teams. tomer, and future). Most of the behavior factors are asso- Defining and aligning the project includes sponsor ciated with more than one outcome measure. We now behaviors that focus on the expected project benefits consider each outcome measure and its associated behav- and success factors. These should be items that would ior factors. help the project team focus on the customer’s real use of the project results. Meeting Agreements Defining the project performance and success The items constituting the meeting agreements factor includes a focus on metrics to measure project success, a include meeting technical specifications while not exceed- strategic consideration of the project’s value, and empow- ing cost and schedule constraints. The first four behavior erment of the project manager to make decisions that he factors (establishing communications and commitment, or she feels are necessary. Each of these is consistent with defining and aligning the project, defining perform- keeping the needs of the project customer (and not just ance/success, and mentoring the project manager) are all the technical specifications) in mind. associated with meeting agreements. Establishing com- Mentoring the project manager includes helping the munications and commitment includes direct sponsor project manager understand the “big picture.” Once again, behaviors, such as personally demonstrating commit- this goes beyond meeting project agreements to realizing ment, and indirect sponsor behaviors, such as ensuring the impact that the project results will have. This is also that communication channels with executives and other consistent with helping the project’s customer to succeed. stakeholders are established and used. Logically, having Prioritizing the project includes sponsor behaviors that strong communications and commitment would seem to ensure that the expected project benefits are understood be consistent with the outcome of meeting agreements and that all stakeholders have been identified. These behav- regarding cost, schedule, and specifications. As a project iors once again reinforce the notion that the project is being runs into inevitable difficulties, having strong commit- performed so that someone (a customer) can effectively use ment on the part of many participants and clear commu- the results. nication channels would appear to be very advantageous. Finally, selecting and establishing the project team is Defining and aligning the project includes direct spon- associated with customer success. This factor includes sor behaviors, such as aligning the project goals with those sponsor behaviors that deal with issues such as team of the firm and aligning project scope with project funding. member skills, knowledge, training, and responsibilities. It also includes indirect sponsor behaviors, such as ensur- It also includes helping the team to establish effective ing that the project goals are clearly defined. The clarity and operating methods, resolve issues, and review project sta- focus these behaviors bring would also seem be consistent tus. The sponsor behaviors comprising this factor are with successfully achieving the agreed-upon project specifi- largely indirect—ensuring that things are done right. The cations subject to constraints of time and money. intent is to develop effective teams that can keep cus- Defining project performance and success includes tomers’ needs in mind as they make decisions. direct sponsor behaviors, such as defining performance In short, the behaviors in these five factors are associ- expectations and empowering others, and indirect spon- ated with the second project outcome—that of helping sor behaviors, such as ensuring that success metrics are the customer to succeed. This is largely consistent with the established. Knowing what is expected certainly appears first outcome of meeting project agreements, but goes to be consistent with meeting cost, schedule, and specifi- beyond. These behaviors both set the broad direction and cation agreements. understanding for the project and help the project man- The final sponsor behavior that is significantly associ- ager and team to become effective. All of these are corre- ated with meeting agreements is that of mentoring the lated with customer success.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 23 The Firm’s Future In summary, there are six sponsor behavior factors that, The last outcome factor is composed of more individual if performed during project initiation, are associated with items than the first two. The firm’s future includes items three project success outcomes. Of those, two—defining such as increasing market share, opening new technologies project performance and success, and mentoring the project and markets, and ultimately achieving commercial success. manager—are associated with all of the outcome measures. The associated sponsor behavior factors are defining project They would appear to be very important behaviors for spon- performance and success, mentoring the project manager, sors. The other four (establishing communications and prioritizing the project, and selecting and establishing commitment, defining and aligning the project, prioritizing project teams. the project, and selecting and establishing the project team) Defining the project performance and success includes are each significantly associated with at least one outcome ensuring that the project’s strategic value is communicated, factor. These six behavior factors and the 41 specific items and defining the project manager’s performance expecta- that comprise the factors should become regular practice for tions. These behaviors can help project participants keep in project sponsors. mind the value of the project to the firm—which is certainly consistent with market share and commercial success. Notes Mentoring the project manager includes helping the 1 New Public Management is a term used to describe dis- project manager understand where the project fits and mon- tinctive new themes, styles, and patterns of public service itoring his or her performance. It would seem that a project management, primarily in Europe. manager’s understanding of how the project is related to the firm’s success, and awareness that her or his performance is References being monitored, could be related to the firm’s success. Association for Project Management. (2005). Glossary Prioritizing the project includes ensuring that project of terms. Retrieved October 29, 2005, from benefits to the firm are studied, the benefits of the project are http://www.apm.org.uk/page.asp?categoryID=4&subCatego emphasized to the steering team, and the steering team pri- ryID=29&pageID=0 oritizes the project. These behaviors are certainly related to a Barki, H., Rivard, S., & Talbot, J. (1993, Fall). Toward an strong awareness of how the project will benefit the firm. assessment of software development risk. Journal of Selecting and establishing project teams includes select- Management Information Systems, 10(2), 203–225. ing team members based upon an understanding of the Center for Quality of Management. (1997). The method skills they need, ensuring that all participants agree on their for priority marking (MPM). Cambridge, MA: Center for specific involvement in writing, and ensuring that a team Quality of Management. quickly resolves any issues that could hinder project per- DeLone, W. H., & McLean, E. R. (1992). Information formance. These sponsor behaviors and others that help systems success: The quest for the dependent variable. develop an effective team can help the firm succeed. Information Systems Research, 3(1), 60–95. Hall, M., Holt, R., & Purchase, D. (2003). Project spon- Conclusions sors under new public management: Lessons from the front- We would like to make two final notes regarding sponsor line. International Journal of Project Management, 21(7), behaviors. First, two of the behavior factors were associated 495–502. with all of the success factors. Defining project performance Helm, J., & Remington, K. (2005). Effective project and success, and mentoring the project manager, might be sponsorship: An evaluation of the role of the executive especially important since each is correlated with all of the sponsor in complex infrastructure projects by senior project success factors. managers. Project Management Journal, 36(3), 51–61. The second comment has to do with what we did not Kloppenborg, T. J., Shriberg, A., & Venkatraman, J. find in our analyses. After starting with 150 sponsor behav- (2003). Project leadership. Vienna, VA: Management iors, we ended up with 41 behaviors that clustered into six Concepts Incorporated. reliable factors that were each correlated with at least one Kloppenborg, T. J., & Tesch, D.B. (2004). Using a proj- outcome factor. There were two additional behavior fac- ect leadership framework to avoid and mitigate Information tors, but neither correlated with any of the success factors. Technology (IT) project risks. In D. P. Slevin, D. I., Cleland, The factor risk planning consists of three behaviors. The & J. K. Pinto, (Eds.), Innovations: Project management research change control factor has five behaviors. The fact that nei- 2004 (pp. 23–34). Newtown Square, PA: Project ther factor is significant in the analysis only means that our Management Institute. respondents do not feel that they are important for the Melymuka, K. (2004a). Surviving the sponsor exit. sponsor to perform during project initiation. They may still be Computerworld, 2/16/2004, 38(7), 40. important during the more detailed project planning stages Melymuka, K. (2004b). Firing your project sponsor. that follow initiation. Computerworld, 2/23/2004, 38(3), 35

24 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Perkins, B. (2005, September). The elusive executive spon- dimension of project management. Project Management Journal, sor. Computerworld, 39(32), 46. 31(1), 6–13. Pinto, J. K. (2004). The elements of project success. In D. I. Schmidt, R., Lyytinen, K., Keil, M., & Cule, P. (2001, Spring). Cleland (Ed.), Field guide to project management. (pp. 14–27). Identifying software project risks: An international Delphi study. Hoboken, NJ: John Wiley & Sons, Inc. Journal of Management Information Systems, 17(4), 5–36. Project Management Institute. (2004). A guide to the project Shenhar, A. J., Tishler, A., Dvir, D., Lipovetsky, S., & Lechler, management body of knowledge (PMBOK® guide) (3rd ed.). T. (2002). Refining the search for project success factors: A mul- Newtown Square, PA: Project Management Institute. tivariate, typological approach. R & D Management, 32(2), Royer, P. S. (2000). Risk management: The undiscovered 111–126.

TIMOTHY J. KLOPPENBORG is a professor of management at Williams College of Business, Xavier University. He holds a PhD in operations management. He is a certified project management professional (PMP), and has authored approximately 60 publications including two books: Managing Project Quality and Project Leadership. Managing Project Quality has been translated into Japanese and Chinese. He is also a syndicated columnist, writing on the topic of project sponsorship. Dr. Kloppenborg is a retired United States Air Force Reserve Officer. His early work experiences were in manufacturing and construction. Mid-career, he took a leave from academia to consult full-time in project management.

DEBORAH TESCH is an associate professor of management information systems at Xavier University. Her current research interests include the role of project participants in project success, system development, and curriculum outcomes assessment. She obtained her DBA in quantitative analysis and management information systems at Louisiana Tech University. She has numerous articles in conference proceedings and journals and a Visual Basic projects text. She is a member of AITP, IACIS, and the Project Management Institute.

CHRIS MANOLIS, PhD, is an associate professor of marketing at Xavier University in Cincinnati. Dr. Manolis’ research has appeared in a number of journals, including the Journal of the Academy of Marketing Science, Journal of Business Research, Journal of Consumer Psychology, Journal of Consumer Marketing, Journal of Services Marketing, Marketing Theory, and Basic and Applied Social Psychology.

MARK HEITKAMP is the vice president of Project Office and joined American Modern in March of 2000 to establish a newly formed project office department within the organization. Prior to this, Mark was vice president of information services at Charter Insurance Companies based in Richardson, TX, USA. His background includes over 25 years insurance industry experience within the project and software development arena. Mark has his PMP from the Project Management Institute along with majoring in computer science at the University of Cincinnati.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 25 ON THE NOVELTY DIMENSION IN PROJECT MANAGEMENT

KLAUS BROCKHOFF, WHU – Otto Beisheim School of Management, Germany

Central Concepts ABSTRACT rojects might be characterized by different degrees of novelty perceived by those who are involved in them. This can influence success, as shown in a Novelty has substantial implications for project P number of contributions (for a summary, see Schlaak, 1999, p. 102). In this management. Because novelty is not objectively determined but perceived by project ini- paper, we draw primarily on earlier empirical research to develop an understand- tiators and participants, diverging perceptions ing of the consequences of different degrees of novelty for project management. At establish first-source management tasks. Also, the same time, we want to introduce results from German studies that have not yet novelty is a multidimensional construct. A been reflected in the non-European scenes of project management practice or the review of dimensions proposed in empirical literature on project management research. measurement of novelty shows that, although about 20 items are used in many studies, no standard measurement approach has yet Project emerged. This hinders comparability and A project is defined in many different ways in the literature. It is not intended here learning from empirical research on project to discuss these differences, sometimes resulting from distinctions in purpose for management. Novelty should be considered as moderating the effects of management on suc- the definitions chosen. To some, it may come as a surprise that, in Germany, even cess, because this is the more encompassing an industry norm sets out to define a project. Since 1980, DIN (German Industry approach. The paper presents empirical Norm) 69901 considers a singular combination of objectives, resources devoted to research results on role structures of project reaching the objectives, duration, separation from other activities, and a specific management, team performance, autonomy of project management and clarity of objectives organization as constituent elements of a project. Separation from other activities as well as ideal versus real characteristics of will become of interest later, when measurement of novelty comes into play. Even project managers. These results are primarily without this particular perspective, it is obvious that in complex, large and fre- taken from studies by German authors with the quently also highly innovative projects, interaction among teams that contribute aim of introducing this research to a broader to one overall goal is mandatory (Hoegl, Weinkauf, & Gemünden, 2004). This audience. Although a substantial wealth of research results is available to support project increases complexity beyond the projects performed by one team, and it differs management, optimisation is not yet possible. from the allocation problem arising from one team serving many projects. It leads But, “one size fits all” is certainly not the right to the differentiation of project-level versus team-level management functions approach, and research results give some indi- (Hoegl & Weinkauf, 2005). cations of how to respond to specific project characteristics. Furthermore, radically innova- In other definitions, the complexity of projects and their “relative novelty” are tive projects call for more specific characteris- mentioned (Frese, 1980). This element, however, is not further specified. It tics of projects managers than for more becomes immediately clear that adding the element of relative novelty to the sophisticated planning aids. aforementioned project definition might lead to difficulties. As in radically new Keywords: novelty dimension; projects, neither objectives, resources, nor duration might find a clear enough fix- multidimensional construct; role structures ation to meet the definition. This is a first indication of the importance of the ©2006 by the Project Management Institute dimension of novelty in project management. Radically new projects, however, are Vol. 37, No. 3, 26-36, ISSN 8756-9728/03 not only difficult to subsume under standard project definitions. These projects also call for management approaches that differ from the routine. As a first requi-

26 AUGUST 2006 PROJECT MANAGEMENT JOURNAL site to choose and apply such rather novel. The reason might be that knowledgeable person may help to approaches, one has to come up with only “they know.” Should “they” be transfer eventual tacit knowledge. This an understanding of novelty. right, the remaining individual could can help the larger number of people catch up by learning. This tells us that involved in learning. Novelty the spectrum of activities necessary for Type 4 projects are those in which— The Subjectivity Issue successful project completion is differ- if not in a very strict sense—“nobody Novelty, newness, or innovativeness, ent from the Type 1 projects: Learning knows.” These are truly innovative which are widely used as synonyms, needs to be included in the project projects, on which everyone involved are at least as difficult to define as the planning and activities. Should “they” agrees. As compared with the Type 1 term project itself. Novelty cannot be be wrong, the project is likely to fail if projects, this might require a different defined in strictly objective ways. the routines applied do not match with management approach. Information Therefore, various “perspectives” the requirements of novelty. This fate, on the choice and application of (Daneels & Kleinschmidt, 2001) are then, is recognized by one person alone. appropriate project management tools considered. However, even one per- Type 3 projects are essentially the is largely missing. Flexibility accompa- spective, namely that of a firm engaged opposite of the Type 2 projects. Except nying project progression is important. in a project, cannot be considered to for one individual, who considers this Evaluations of intermediate results are be homogeneous. Rather, because nov- project type not to be rather novel, all necessary as a basis for continuation or elty is imagined, it is a subjective phe- others lack the special knowledge that termination decisions. nomenon. Excluding one-person the one individual seems to have. “I One of the implications that the projects, the definition derives from know,” this person could say. Lotka’s four project types have is for project the imaginations of a group of people Law (1926) describing the skewness of planning. If novelty perceptions who are related as contributors to a expertise distribution among a group diverge, preferred planning modes will project. Because separation of project of people gives an explanation for the diverge as well for the different people activities from other activities is part of divergent perceptions of novelty. involved. This could become a source the project’s definition, this group of However, the problem for project man- of conflict. Problems of a similar people should be identified. Only if agement is that the disagreeing person nature might arise if multiple team perceptions of all these individuals are might either be a genius or a swindler. projects are considered in which each identical might we speak of an objecti- It would be nice to have signals, such team perceives its task as involving a fied notion of novelty. However, such as exams, papers, references etc., to tell different degree of novelty. Securing agreement cannot be assumed as a one from the other. But even signals cooperation among the teams to rule. By abstraction, let us assume that might not be foolproof. Johann enhance team innovation (Tjosvold, a group of people related to a project Kunckel (1630 to 1703) was a chemist Tang, & West, 2004) might be difficult can be partitioned into one individual who, among other things, successfully to achieve. and the remaining members of the completed projects to produce ruby Because novelty evaluations such group, where the individual might glass in large quantities, and who re- as in Figure 1 apply to a specific point agree or disagree with the rest of the discovered phosphorus. He wrote a in time, project progress will lead to a group with respect to the novelty of a number of well-received books, in par- change of the respective evaluations particular project. It is not uncommon ticular on glass making, but never (Lange, 1993, p. 125 et seq.). There are to find a powerful project promoter to received an academic education. indications that this calls for phase- assume that a project is not really Obviously, he had promised to the specific project management (Hoegl & novel, while experts despair of the level Saxon king that he was also able to Weinkauf, 2005) or, putting this differ- of novelty they encounter. A similar produce gold. After the king’s death, he ently, novelty-specific project manage- observation with respect to top man- fought for outstanding remuneration. ment. Although both technical agement as the promoter was made by However, the chancellery told him that uncertainty and market uncertainty are Green (1995). This leads to four major “if He could produce gold He does not reduced in successful novel product types of projects (see Figure 1). need a salary; but if He cannot produce development projects, market uncer- Type 1 projects are routine projects. gold, why then should He be remuner- tainty is not reduced as frequently in All people involved agree on the zero- ated?” With respect to this project type, unsuccessful projects (Lange, 1993). to-low degree of novelty. With respect to supervision might therefore be of pri- The project types sketched in the knowledge necessary to perform the mary importance. Defining milestones Figure 1 have another important impli- projects of this type, one might say that and checking on project performance cation, namely for empirical research “we know” how to do it. Standardized are mandatory activities. Incentives or as well as for reporting. The potential project planning can be applied. controls that prevent opportunism disagreement by project contributors Type 2 projects are considered rou- resulting from the asymmetric distri- on the evaluation of novelty tends to tine by all but one individual (or a bution of knowledge are mandatory. produce substantial single-informant minority of a few). This individual Documentation of the project progress errors in empirical research. Therefore, considers the respective project to be and the cooperation of many with the data should be collected from many

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 27 ationalized in a way that displays the High adequacy or fit of technological resources, the remaining two con- Type 3 Type 4 structs cut across the fit and familiarity Genius or Swindler Innovation dimensions chosen by Daneels and “I know” “Nobody knows” Kleinschmidt (2001). (3) Schlaak (1999) substantially broadened these concepts. Starting out from Leavitt’s (1965) four dimensions of organizational change—task, struc-

Objective Novelty Type 1 Type 2 ture, technology, and actors—he Routine Learning expands technology into a resources Low “We know” “They know” construct, and substitutes processes for the actors’ dimension. These constructs are operationalized by 40 items, 20 of Low High which can be related to the familiarity Subjective Novelty construct and another 17 to the fit construct. Careful analysis of a set of proj- Figure 1: Quasi-objective or subjective perceptions of novelty and project types ect data narrows the number of items to 24, grouped into seven factors. Three informants in order to control this indicated via one-dimensional of these represent the fit, while four error. This is of particular interest for scales—to the multivariable approach- represent familiarity. Consequently, Type 2 projects and Type 3 projects. In es of today (for instance see Daneels & Schlaak (1999) argues that his the most encompassing review of new Kleinschmidt, 2001, p. 359; Garcia & approach is different from that of product success studies to date, Ernst Calantone, 2002; Hauschildt, 2004, p. Green et al. (1995). However, the seven (2001, 2002) identified only two out 14; Schlaak, 1999, p. 91). With respect factors used to measure novelty reach of 51 studies that explicitly and sepa- to the latter approach, four contribu- far beyond the earlier two dimensions rately studied the hierarchical or functions are of major importance. These of market and technology. In fact, they tional levels of respondents as a source are very shortly reviewed, keeping the point at a strategic issue of the project of error. In his own study, he finds that definitional problems in mind. definition. As already mentioned, this the informant bias is responsible for (1) Green, Gavin, & Aiman-Smith includes “separation from other activi- roughly 30% of the total variance in (1995) considered four constructs to ties,” mostly meaning other projects product success measures. The same define novelty characteristics: techno- performed in parallel. Schlaak (1999) author makes it clear that indirect as logical uncertainty, technical inexperi- showed how a project can have sub- well as direct indications show sub- ence, technology cost, and business stantial influence on the whole organi- stantial respondent errors with respect inexperience. These are operational- zation into which it is embedded. Even to variables that are associated with ized by 17 items, one of which has a product innovation project can lead novelty in many of the ways in which been eliminated for statistical reasons. to novel procurement and production this is measured. These different Although the first of the constructs, processes, and to changes in the infor- approaches of measurement need to technological uncertainty, offers an mal or formal organization. In this be explained: almost objective view of how the sci- sense, there is no separation from other entific community is able to handle activities, particularly if high degrees of Measuring Novelty the project, the remaining three items novelty are observed. Such emanation Novelty has no meaning by itself, but refer to the performing firm alone. effects can be a source of opposition only in relation to some characteristic. (2) Daneels and Kleinschmidt against projects, because not everybody Even if we limit further considerations (2001) used 19 items to operationalize loves innovation. Explicit considera- to the Type 1 and Type 4 projects, the four constructs: familiarity of the firm tion of emanation effects might move characteristic to which novelty relates considered with the technology and a project from Types 1 or 2 to Types 3 needs to be made explicit. This is one familiarity with the market, as well as or 4. The research by Schlaak (1999) subject of measuring novelty. Even if the fit or adequacy of the firm’s also casts doubt on the proposition of we limit ourselves to product innova- resources in the marketplace, and with independence of the familiarity and fit tion projects, this field has produced a respect to its technological capabilities dimensions. As observed in many fac- substantial number of approaches, to achieve the project objective. This tor analysis studies, the seven factors ranging from the earlier and simple firm-specific approach has some over- he identifies are not completely inde- “new/not new” dichotomies—which, lap with the firm-specific items of pendent of each other. This could in their most simple form, do not even Green et al. (1995). However, except mean that fit and familiarity are not address which novelty characteristic is for “technological cost,” which is oper- totally independent of each other. For

28 AUGUST 2006 PROJECT MANAGEMENT JOURNAL instance, novelty to the sales market is acteristics in measuring novelty. Multi- increment. Thus, in Equation 4 an a familiarity dimension, while the fit informant approaches are called for, additional parameter (e) is estimated. dimension is represented by resource but are not common. These could help Although this consumes another costs in R&D, production, and market- to control for single informant errors. degree of freedom, it adds explanatory ing. Both are correlated at a lower Perceptions matter, as can be seen by power if e is estimated as significantly level. Similarly, the “higher order” fac- referring to the project categories of different from 0. An even more tor of technology and production is Figure 1. expanded approach is represented by a composed of two familiarity factors A standard measurement proce- combination of Equations 2 and 4, and one fit factor. dure would establish an important which leads to Equation 5: (4) An even broader perspective step toward better comparison of the with respect to the consequences of results of empirical studies, and it y = c’ + x(e + fz) + d’z (5) project novelty is taken by the meas- would help management to better clas- urement approach of Gemünden and sify their projects. with d’ = db. Because plausibility, occa- Salomo (2005). Their four dimensions sionally supported by scatter plots, has of novelty relate to the market, the How Novelty Works: The Principles it that some independent variables can technology, the organization into of Influence have nonlinear effects on the depend- which the project is embedded, and Another important aspect is that of ent variable beyond the multiplication the external environment as defined by whether novelty is a moderator or an of x and z, linear approaches can be regulatory or societal aspects. Twenty independent variable (Daneels & overly simplistic. Comparing a nonlin- items is used for construct definitions. Kleinschmidt, 2001). These can be ear expansion of Equation 1 with an Again, familiarity and fit cut across the considered as two different principles approach such as Equations 4 or 5 four constructs. The broadening of the of influence that novelty may have on might therefore be of interest. perspective is not only manifest in the projects. Assuming strictly linear rela- From this short presentation, it is environmental construct, but also, for tionships, and using {y, x, z} as vari- concluded that further empirical instance, in the market aspects of nov- ables and {a, b,…, f} as parameters, research should adopt the moderator elty. The authors ask project represen- the two principles can be illustrated principle in trying to explain influ- tatives to what degree they think that very easily. The variable y indicates a ences of novelty on project manage- customers need to change their behav- particular project outcome, such as ment. Furthermore, nonlinearities ior and attitude to appreciate the proj- one dimension of project success or a should attract more attention than ect outcome. Other than originally composite measure of project success. they have to date. imagined, the perspective taken in the The variable x indicates a project man- evaluation is not represented by a agement characteristic, while z is a Novelty Effects on Project Management member of the group who holds a spe- measure of novelty. The latter implies a The following presents selected empir- cific perspective similar to that of the characteristic, which could also be rep- ical findings of effects of novelty on firm hosting the project. resented by a vector with respective project management. This has two What can be learned from this? consequences for the parameters. Error major limitations, besides the inability Even though great progress has been terms are neglected here. Then, novelty to cover all studies. First, the measure- made in measuring novelty, no stan- as an independent variable is observed ment issues previously raised limit dard has yet been established. This is in the following equations: comparisons. Second, we draw on even more disturbing as the approach- studies that advance the hypothesis of es previously referred to focus only on x = a + bz (1) a causality that is mostly tested by non- product innovation projects. Novelty y = c + dx = c + d(a + bz) (2) causal approaches. Thus, the possibili- with respect to other types of projects ty of project management influencing should be measured by the same In Equation 2, z is simply substi- novelty rather than the other way methodological rigor. The broadening tuted for x. As a moderator, novelty is around is not explicitly covered (see of the novelty perspective adds more modeled in the following equations: Figure 2). The dotted line shows this realism to the evaluation, but at the causality. However, this relationship same time might be an additional d = e + fz (3) does not appear to exist. Rather, both source of variance. y = c’ + x(e + fz) (4) project management and novelty Plausible concepts of novelty are might be determined by underlying not easily replicated in follow-on stud- It is easily seen that Equations 2 factors, such as the governance of the ies. Constructs cannot be considered and 4 are conceptually different, and organization into which the project is generally valid, although authors tend also that Equation 4 is the more gener- embedded. In Equations 1 through 4, to use about 20 items for their meas- al approach. In empirical estimation of this is implicit in a lack of time indices urement and at least some basic agree- unknown parameters, it will not be at the variables. The bold lines and the ment exists as to the necessity of possible to differentiate between c’ and dashed line show the relations mod- including technology and market char- c + da, both being considered as an eled in Equation 5.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 29 project management influences project success only via its influence on the Project Management (x) teams. For the most novel projects standard instruments of project management play a very reduced role. It is (4) Novelty (z) suggested that these projects require considerable new knowledge, which consequently increases the importance of the team for project success. Also, Project Success (y) d’z because these projects are relatively complex and consume more resources than any other type of project, the role Figure 2: The causality issues between project management and novelty of the project management is stressed. The slight reduction of the importance In the following, three aspects purchases of computers in companies. of top management’s role is explained shall be considered: project manage- He identified a sponsor’s role (called by the assumption that this type of ment structure, clarity of project objec- power promoter) and an expert’s role project is so prominent within the tives and autonomy of project as the two important roles that need to organization that it does not need as management, and the project manager. be present and represented at the same much top management attention as time to enhance the chance of project might be necessary to keep a less Project Management Structure success. This research was expanded prominent project on a path to success. Project management structure reflects into integrating the role of the project In this study, an overly simplis- a hierarchy of top management, with champion (process promoter), and tic—namely, static—view is presented. project manager and project team as more recently to the gatekeeper for the In reality, the roles of promoters are the acting persons. Involvement of top transfer of knowledge as well as the evaluated differently with respect to management, competencies assigned relationship promoter to support proj- different project phases. The people to the project manager, and expertise ect relations with other organizations representing different roles can change, of the project team have positive influ- beyond the project’s host (Chakrabarti and the relevance of certain roles for ences on project success. Lechler & Hauschildt, 1989; Hauschildt & project success can vary with novelty (1999) presented an overview of the Gemünden, 1999). The promoter roles (Folkerts, 2001). In particular, if spon- results that support this. The same are associated with particular hierar- sors’ support in highly novel projects is author shows in his LISREL study of chies of their representatives as well as discontinued, these projects tend to 257 successful and 191 unsuccessful with particular resources that their rep- fail. Furthermore, the interaction or projects that the total affect on project resentatives can mobilize for project cooperation of the three role bearers is success of these individuals or groups success. A recurrent result of empirical not studied. In addition, the inclusion of people can vary with project type. research is that the roles should all be of the novelty dimension is a rather Three types of projects identified by represented in project management crude one. Both of these aspects are Lechler (1999) exhibit—among other and that cooperation among the repre- covered in a much more elaborate characteristics—increasing “innova- sentatives of these roles is decisive for study in which the novelty dimension tiveness” and “technological risks.” project success. is the major issue (Papies, 2005). At This observation helps in integrating a If one accepts Lechler’s association the same time, this study looks at the stream of research on the management of people and roles, it is possible to same projects in different phases of structure for innovative projects, which study the importance of each of the their development. was not originally integrated with the three roles for project success (Table 1). With respect to novelty, Papies project management literature. In this Here, this is of particular interest with (2005), with some restrictions, follows descriptive, rather than normative, lit- respect to the novelty dimension. For the previously-mentioned concept of erature, the roles of sponsor, champion both, the project manager (champion) Schlaak (1999). Furthermore, he and expert together with their cooper- and the project team (experts), their looked at the same new product devel- ation are identified as crucial for proj- affect on project success increases with opment projects in the three phases of ect success. Lechler (1999) associates the two novelty variables. The effect of concept development, advanced devel- top management with the role of top management is almost identical, opment, and testing or market intro- sponsor, the team with the role of the with a slight reduction of influence duction. In the first two phases, experts, and project management with on success for the most innovative interactions of the roles of sponsor, the role of the champion. projects. The top management (spon- champion and expert are studied The role model was originally sor) influences success directly as together with top management sup- adopted by Witte (1973). He studied well as indirectly via its influence on port and cooperation of the promoter’s projects set up to manage first-time project management and teams. The roles. The sponsor, the champion and

30 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Project Type A Project Type B Project Type C effectiveness. Otherwise, insignificant (n=192) (n=120) (n=102) results are noted for this phase of project development. Novelty Low Medium High In the advanced development phase, Risk Low Medium High it is notable that higher degrees of nov- Project size Medium Small Large elty have a negative influence on efficiency in all cases. Obviously, high Total impact of top .60 .64 .59 degrees of novelty are correlated with management high degrees of uncertainty, which in Total impact of .12 .16 .23 turn lead to surprises, spoiled well- project management laid-out plans, etc. The moderating Total impact of .41 .40 .63 effect is similar to that of the preceding project team phase with respect to top management

Source: Lechler (1999, pp. 205 & 207) support. It is even stronger with respect Table 1: Impact on project success by three types of role representatives and project characteristics to the influence of a sponsor or an expert on USP. However, effectiveness the expert roles are all identified and bution to the project’s USP. The moder- of team quality and cooperation, as considered with respect to their influ- ating effect of novelty (represented by well as efficiency of team quality, are ences on the success variables. Thus, a the parameter f in Equation 5) is signif- all negatively influenced by higher very elaborate approach is chosen. icant in only a few situations. novelty during this phase. This may Novelty is considered as a moderator, In the conceptual phase of the proj- result from the weakness of team man- as in Equation 5. ects studied, a significantly positive agement in dealing with a situation as In Tables 2 and 3, a summary of moderating effect occurs five times. complicated as that of high novelty. results is presented. In these tables, – or This means that, although novelty In software development, project + signs represent a significant parame- might reduce the value of the success team performance can be moderated ter estimation in the respective direc- variables considered, this reduction by novelty of the projects assigned to a tion, while a 0 stands for an can be more than compensated for by team. Interestingly, high levels of nov- insignificant result. It is found that in higher degrees of novelty if an expert is elty impact on the relationship the two relevant project phases, novelty involved or if cooperation among between team quality and efficiency, itself (as measured by the parameter d’ sponsors is ensured, or if top manage- but do not significantly impact on the in Equation 5) is perceived as having ment lends its support, or if team qual- relationship between team quality and mostly significant negative effects on ity can be assured. In four of these five effectiveness as seen by team leaders project efficiency and project effective- cases, this occurs with respect to effi- and managers; team member ratings ness, but positive effects on the contri- ciency, and only once with respect to are non-significant overall (Hoegl, Parboteeah, & Gemünden, 2003). Non-significant negative impact of Phase Success Promoter Type / d’ f higher novelty levels on team quality Measure (y) Variable (x) can be compensated by the moderat- Concept Effectiveness Sponsor - 0 ing effect. This result seems to contra- Development Efficiency Sponsor - 0 dict earlier findings, and suggests USP Sponsor + 0 managing team quality in response to Effectiveness Champion - 0 novelty rather than using unified Efficiency Champion - 0 approaches across the board. USP Champion + 0 Effectiveness Expert - + Clarity of Project Objectives and Efficiency Expert - + USP Expert 0 0 Autonomy of Project Management Effectiveness Cooperation - 0 Some of the negative direct influences of Efficiency Cooperation - + high novelty on efficiency and effective- USP Cooperation 0 0 ness of projects might be related to a lack Effectiveness Top man. support - 0 of structuring objectives in such projects. Efficiency Top man. support - + Prescriptive literature suggests that proj- USP Top man. support + 0 ect objectives should be non-complex, Effectiveness Team quality - 0 measurable, specific, tangible and easily Efficiency Team quality - + verified, among other characteristics USP Team quality + 0 (Kerzner, 1984, p. 344). Primarily, this Source: Own construction according to results by Papies (2005) seems to apply to Type 1 projects (see Table 2: Novelty as moderator in structuring project management in the concept development phase Figure 1). There is nothing wrong with

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 31 Phase Success Promoter Type / d’ f (Krieger, 2005). Increasing autonomy Measure (y) Variable (x) of the project management can lead to cutting ties with the originally support- Concept Effectiveness Sponsor 0 0 ing organization together with Development Efficiency Sponsor - 0 insignificant effects of the novelty USP Sponsor + + moderator. Thus, for instance, recourse Effectiveness Champion 0 0 Efficiency Champion - 0 on knowledge or other resources might USP Champion + 0 be cut off. Face-to-face interaction as Effectiveness Expert 0 0 an indicator of social autonomy of a Efficiency Expert - 0 project is, however, significantly posi- USP Expert + + tively associated with success, and the Effectiveness Cooperation 0 - novelty moderator has a positive influ- Efficiency Cooperation - 0 ence. This calls for co-locating teams USP Cooperation + 0 that work on highly innovative proj- Effectiveness Top man. support - + ects within the organization to benefit Efficiency Top man. support - + from the richness of personal informa- USP Top man. support + 0 tion exchange and an easier way of Effectiveness Team quality 0 - Efficiency Team quality - - motivating identification of team USP Team quality + 0 members with the project. With respect to another autono- Source: Own construction according to results by Papies (2005) my dimension, namely autonomy in Table 3: Novelty as moderator in structuring project management in the advanced development phase formulating objectives, it is found that too much autonomy might para- trying to achieve these characteristics, aged if project managers enjoy a high lyze the synergies that the supporting even in novel projects (of Type 4 in level of autonomy with respect to orga- organization, hopes for when starting Figure 1). However, such objectives nizational structure and physical sepa- the project (Brockhoff & Schmaul, should not be upheld irrespective of ration from routine operations, 1996). A loss of focus can be the the path of the knowledge accumula- resource availability and use (Krieger, result. Even the Internet pages that tion in the project. Case studies show 2005), as well as social autonomy. The report on project failures are abun- that this request, while naively natural, latter is a short description for either dant with not only the usual criticism is frequently not met, particularly if offering personal, face-to-face coopera- of lack of resources, weakness of proj- commitments among different con- tion among team members or for team ect management or failure of top tributors to a project have to be agreed members remaining in their home management support, but also with upon (Hauschildt & Pulczynski, 1992; institutions with only virtual coopera- respect to overambitious goals, proj- Hauschildt & Pearson, 1994). Penalty tion (Gemünden & Salomo, 2005). A ect objectives driving away from the for breach of contract at the institu- fine overview of these suggestions and vision of the supporting organization tional level, or face-saving at the indi- findings is given by Krieger (2005, p. or goal changes by top management vidual level, might work toward 37). Another quite similar expression without proper communication (for inflexibility, but also, unfortunately, for this suggestion is the high-powered example, Hedman, 2005; Rossi, toward project failure. To achieve flexi- project management (Wheelwright & 2005). Again, the novelty dimension bility in project management without Clark, 1992). may come into play as a moderating totally losing sight of the original As before, the novelty dimension factor. Because some projects are driv- objectives, a particular set of character- can be studied as moderating possible en by the availability of resources istics of the project manager or the direct relationships between variables (means), including knowledge, and champion could be called for, or a cer- measuring various dimensions of other are driven by the availability of tain level of autonomy in managing autonomy and project success. This a clear view of purposes (ends) (for the project. Results on both of these casts doubts on earlier findings of the means-end-classification of inno- issues can be presented. extending autonomy in projects with vative projects, see Hauschildt & As mentioned before, separation increasing novelty to achieve project Pearson, 1994), the change of objec- from other activities is considered a success (see for instance, Christensen & tives can have substantially different definitional characteristic of projects. Overdorf, 2000; Gerwin & Moffat, success influences. It is quite plausi- This could help to achieve the level of 1997; Simon, Houghton, & Gurney, ble to assume that such changes can flexibility needed for successful high- 1999). Several arguments can be be helpful and supportive for novelty projects. The same idea seems advanced to explain findings that do resource-driven projects, while it can to have led quite a number of not show positive effects of structural kill purpose-driven projects, unless researchers to suggest that a high autonomy and limited effects of the purpose itself undergoes respec- degree of novelty might best be man- resource autonomy on success tive parallel changes.

32 AUGUST 2006 PROJECT MANAGEMENT JOURNAL The findings call for a complicated tion is another interesting question. Characteristic Mean multidimensional optimization of Even a casual look at the project man- Weight autonomy as a means to achieve the agement literature reveals the extreme- flexibility necessary when novelty ly demanding set of characteristics that Past development 20.8 experience increases. First, the optimization has to the “ideal” project manager should be achieved with respect to the novelty meet. As previously speculated, this Systematic and analytic 18.4 thinking dimension, where we have indications may be even more important in the that with respect to the market and case of high-novelty projects because Ability to plan and organize 24.7 technology dimensions more autono- the multidimensional optimization Ability to motivate 17.6 my can be allowed as compared with has to be negotiated with the manage- Creativity 18.5 the organizational and environmental ment of the supporting organization, fit dimensions. This can be concluded and because maintaining the optimal Source: Keim (1997, p. 234) from the observation that the level of degree of internal flexibility—for Table 4: Mean weights for characteristics of “ideal” R&D project managers market or technological novelty in the instance, with respect to the project sense of fit does not significantly corre- objectives—is very demanding. ture on project manager characteristics late with success measures, while some Indeed, the accumulated characteris- to describe reality (Keim, 1997, p. 151). negative correlations can be observed tics that practitioners and the academ- By employing factor analysis, these with respect to organizational and ic project management literature use to were assigned to seven factors, clus- environmental novelty dimensions. define “ideal” project managers has led tered to identify five types of project These come close to a familiarity one researcher to speak of the ideal managers. These, in turn, were related dimension. The management task here project managers as “heroes” in the to project success. Success was mainly was nicely phrased by Yasutsugu classical sense of the word or even measured as technical success, but all Takeda, then head of research at “archangels” (Drumm, 1996). projects had already been introduced Hitachi Corp. He did not like “blue Friedman, Fleishman, and Fletcher into the markets, some showing a suc- sky” research projects, but rather (1992), looking at R&D projects, ques- cessful market life of three years. Again “north star” projects: almost equally tioned the necessity of technological following the analysis by Keim (1997), removed from the present technology qualifications for project managers Table 5 summarized her results. and market, but organizationally and favor their managerial competen- From Table 5 we can conclude that much better focused. Second, the opti- cies to achieve success. A similar indi- a project manager who interacts with mization has to be achieved with cation, particularly from the point of his or her team, who can motivate and respect to the dimension of autonomy, view of Japanese managers as com- who can critically or logically evaluate where issues of organizational struc- pared with German or U.S. managers, project performance, achieves the best ture, resource availability, and freedom supports this view (Brockhoff, 1990, project results. Positive, but not out- to dispose of resources as well as social p. 87), although with no particular ref- standing, values with respect to the autonomy have to be considered. erence to projects. By application of other characteristics certainly help to Third, optimization has to be achieved conjoint analysis, Keim (1997, p. 219) do such an outstanding job. This char- with respect to the success dimension identified the relative weight of com- acterization is particularly interesting if and the phase of project work, which petencies requested for R&D managers confronted with the results for the might serve as signals for the ultimate (see Table 4). It is apparent that the group of the least successful project but hardly controlled financial success past experience on the job, together managers. They stand out with respect in the market. with systematic and analytical think- to their problem-solving capabilities, It is obvious that the project man- ing, counts less than the combination but they miss out on all other charac- agement knowledge to date does not of creativity, motivation, and the abili- teristics. One can imagine that this offer enough information for such an ties to plan and organize. The highest characterizes managers who are used optimization. This is particularly so, level of each characteristic is always to exercising power, to put high pres- because the study by Krieger (2005), preferred to lower levels, with one sure on a team with little regard for the which opened this presentation of the exception. For creativity of the project specific difficulties of highly uncertain optimization problem was based on manager, both the lowest and highest and novel jobs. It was mentioned 104 relatively radical or very novel types level receive almost equal weight, before that the dominance of a power of product innovation projects. To while medium levels find small num- promoter in novel projects is not a what degree the results can be corrob- bers of supporters. favorite driver of success. Type 2 repre- orated with respect to other types of In reality, hardly any project man- sents the most frequently observed projects, we do not know. ager will be able to live up to the high- project manager. This manager comes est expectations with respect to each close to the mean of characteristics of Project Manager and every one of the characteristics all project managers with respect to To what degree an able project manag- mentioned in Table 4. Therefore, 24 five out of the seven characteristics, but er can make good for missed optimiza- items were identified from the litera- with lowest values for two of them.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 33 Characteristics Project Manager Project Manager Project Manager Project Manager Project Manager Type 1 Type 2 Type 3 Type 4 Type 5 Can interact with team 1.47 0.12 -0.47 -1.06 -0.15 Has business experience 0.23 0.04 0.92 -0.57 -1.01 Has problem-solving 0.49 -0.55 -0.15 0.07 1.33 capability Shows initiative and 0.49 -0.62 0.95 0.68 -1.25 creativity Can learn and organize 0.12 -0.03 -0.99 0.47 -0.08 Can motivate and cooperate 0.62 -0.35 -0.0 0.40 -0.21 Reflective, logical thinker 0.33 -0.22 -1.15 0.32 -0.48 Relative share of project 20% 38% 11% 20% 11% manager types

Project success scale values 5.6 4.8 4.7 4.5 3.8

Source: According to Keim (1997, pp. 160–161, & 189) Characteristics of project managers are evaluated by T-values, based on standardized factor values. They can be interpreted as deviations of one cluster of project managers from the overall mean. Extreme positive and negative T-values per each characteristic are highlighted for better reading. Table 5: Realistic types of R&D project managers, their frequency of appearance, and success

One can imagine that working under entiated with respect to the degree of study. We are critical of such associative, the leadership of this type of manager novelty and novelty characteristics. not empirically tested, reasoning, par- means to live with a certain degree of Second, this suggests that the applica- ticularly because the low levels of learn- chaos, but the average business experi- tion of more exacting practices and ing and reflective abilities of Type 3 ence and the average learning ability methods needs particular project man- project managers cast doubt on Keim’s might still create acceptable results. The ager abilities or characteristics. This (1997) conclusions. It would not be wheeler-dealers of Type 3 have to fight could well add a third dimension to surprising if—in further, more expand- against their lack of other characteris- Table 5. Rubenstein, Chakrabarti, ed, analyses—one would find that, for tics, specifically learning and more O’Keefe, Souder, & Young (1976, p. 18) instance, Type 5 is the sponsor of a pet reflective responses to problems that once said that one should not believe project that he or she personally man- arise. Type 4, finally, seems to lack expe- that “organization structure, control ages. Type 4 has a number of charac- rience, which cannot be fully compen- mechanisms, formal decision-making teristics that could more convincingly sated for by their learning, initiative, processes, delegation of authority, and characterize them as experts. motivation, and reflection. This type other formal aspects of a so-called well- Furthermore, no information is seems to lack a frame of reference for run company are sufficient conditions available on the cooperation among the positively marked characteristics. for successful technological innova- the team members. Thus, the extant Although these results offer a great tion.” In view of the results in Table 5, work is no true test of the promoter deal of insight into the characteristics one might say that these instruments model and its hypotheses. In all fair- of project managers for novel projects, are necessary, but they need to be ness, it was not planned as such a test. it is unclear whether they can be applied by project managers who know applied to other project types. Also, when to choose which instrument, and Conclusion because all of the projects considered to determine the level of its application In establishing efficient and effective in the study had been introduced into if this is possible beyond a simple alter- project management, a multidimen- the market, no major project failures native of choice. sional optimization problem has to be are included in the study. Their inclu- In a separate part of her study, solved. One of the driving forces to be sion might sharpen the picture that Keim (1997, p. 214) discussed the observed in striving for a solution is could be drawn, or introduce a new question of associating the characteris- the moderating effect of novelty, which type of project manager who scores tics of the five types of project man- is itself a multidimensional concept. It below average on all characteristics. agers can be associated with the is difficult to assess novelty, because it Souder and Jenssen (1999, p. 198) promoters. She concluded that Type 1 is a subjectively held concept related to concluded that “more exacting new can be associated with the champion dimensions that can be company- product development practices are and Type 3 has a profile that corre- internal or -external. No standard required to achieve success in unfamil- sponds with the expert. This follows measurement approach has yet iar market environments.” First, this the same associative reasoning as is evolved. In fact, the advance of meas- means that methods should be differ- already known from Lechler’s (1999) urement concepts is not without con-

34 AUGUST 2006 PROJECT MANAGEMENT JOURNAL flicts. This limits comparisons across Drumm, H. J. (1996). Das perspective. IEEE Transactions of studies. Furthermore, the way that nov- Paradigma der Neuen Dezentralisation Engineering Management, 42(3), elty is empirically modeled can have [The Paradigm of new 223–232. influences on results with respect to Decentralization.]. Die Betriebswirtschaft, Green, S. G., Gavin, M. B., & the success dimension of projects. It is 56(1), 7–20. Aiman-Smith, L. (1995). Assessing a shown that the approach to consider Ernst, H. (2001). Erfolgsfaktoren multidimensional measure of radical novelty as a moderator is more general neuer Produkte. Grundlagen für eine technology innovation. IEEE than the approach to consider it as an valide empirische Forschung [Success fac- Transactions on Engineering Management, immediate success variable alone. tors of new products: Foundations of 42(3), 203–214. One of the contributions of the valid empirical research]. Wiesbaden, Hauschildt, J. (2004). present paper is to present a few results Germany: Gabler Verlag, Deutscher Innovationsmanagement [Innovation of empirical research on project man- Universitätsverlag. management]. München, Germany: agement done in Germany, and to Ernst, H. (2002). Success factors of Vahlen Verlag. relate this to some of the issues of proj- new product development: A review of Hauschildt, J., & Gemünden, H. ect management that are discussed the empirical literature. International (Eds). (1999). Promotoren. Champions internationally. Here, concepts as that Journal of Management Reviews, 4(1), der Innovation [Promotors. Champions of the promoters of project manage- 1–40. of innovation]. Wiesbaden, Germany: ment might prove to be fruitful in Folkerts, L. (2001). Promotoren Gabler Verlag. other research as well. The moderating in Innovationsprozessen. Empirische Hauschildt, J., & Pearson, A. W. influence of novelty can be combined Untersuchung zur personellen (1994). Ends, means and innovation with the promoter concept to explain Dynamik [Promotors in innovation management. Creativity and Innovation project success. It can also be used to processes. Empirical research of the Management, 3(3), 162–166. identify more formal project manage- human resource dynamics]. Hauschildt, J., & Pulczynski, J. ment methods as being adequate for Wiesbaden, Germany: Deutscher (1992). Rigidität oder Flexibilität der the less novel projects, while certain Universitätsverlag. Zielbildung in Innovationsprozessen? characteristics of project managers Frese, E. (1980). Projektorganisation [Rigidity and flexibility of goal forma- might be more important than project [Project organization]. In Handwörterbuch tion in innovation processes]. management methods for achieving der Organisation (2nd ed.) (pp. Zeitschrift für Organisation, 61(2), success of highly innovative projects. 1960–1974). Stuttgart, Germany: C. E. 74–81. Poeschel Verlag. Hedman, E. R. (2005). Why tech- References Friedman, L., Fleishman, E. A., & nology projects fail. Retrieved Brockhoff, K. (1990). Stärken und Fletcher, J. (1992). Cognitive and inter- September 7, 2005, from www.thes- Schwächen industrieller Forschung und personal abilities related to the pri- pacereview.com/article/372/1. Entwicklung [Strength and weaknesses mary activities of R&D managers. Hoegl, M., Parboteeah, P., & of research and development]. Journal of Engineering and Technology Gemünden, H. G. (2003). When team- Stuttgart, Germany: C. E. Poeschel Management, 9(3), 211–242. work really matters: Task innovative- Verlag. Gemünden, H. G., & Salomo, S. ness as a moderator of the Brockhoff, K., & Schmaul, B. (2005). Der Einfluss der teamwork-performance relationship in (1996). Organization, autonomy, and Projektautonomie auf den software development projects. Journal success of internationally dispersed R&D Projekterfolg [The influence of project of Engineering and Technology facilities. IEEE Transactions on autonomy on success]. In J. Management, 20(3), 281–302. Engineering Management, 43(1), 33–40. Amelingmeyer & P. E. Harland (Eds.), Hoegl, M., & Weinkauf, K. (2005). Chakrabarti, A., & Hauschildt, J. Technologiemanagement & Marketing – Managing task interdependencies in (1989). The division of labour in innova- Herausforderungen eines integrierten multi-team projects: A longitudinal tion management. R&D Management, Innovationsmanagements (pp. 63–82). study. Journal of Management Studies, 19(1),161–171. Wiesbaden, Germany: Gabler Verlag. 42(6), 1287–1308. Christensen, C. M., & Overdorf, M. Garcia, R., & Calantone, R. (2002). A Hoegl, M., Weinkauf, K., & (2000). Meeting the challenge of dis- critical look at technological innovation Gemüenden, H. (2004). Inter-team ruptive change. Harvard Business Review, typology and innovativeness terminolo- coordination, project commitment, 78(2), 67–76. gy: A literature review. Journal of Product and teamwork in multi-team R&D Daneels, E., & Kleinschmidt, E. J. Innovation Management, 19(2), 110–132. projects. Organization Science, 15(1), (2001). Product innovativeness from Gerwin, D., & Moffat, L. (1997). 38–55. the firm’s perspective: Its dimensions Withdrawal of team authority during Keim, G. (1997). Projektleiter in der and their relation with project selection concurrent engineering. Management industriellen Forschung und Entwicklung and performance. Journal of Product Science, 43(9), 1275–1287. [Project managers in industrial Innovation Management, 18(6), Green, S. G. (1995). Top manage- research and development]. 357–373. ment support of R&D projects: A strategic Wiesbaden, Germany: Deutscher

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 35 Universitätsverlag. Gabler Verlag. Simon, M., Houghton, S. M., & Kerzner, H. (1984). Project man- Lotka, A. J. (1926). The frequency Gurney, J. (1999). Succeeding at inter- agement (2nd ed.). New York: Wiley. distribution of scientific productivity. national corporate venturing—Roles Krieger, A. (2005). Erfolgreiches Journal of the Washington Academy of needed to balance autonomy and con- Management radikaler Innovationen. Sciences, 16(6), 317–323. trol. Journal of Applied Management Autonomie als Schlüsselvariable [Successful Papies, S. (2005). Prozessbegleitende Sciences, 8(2), 145–149. management of radical innovations. Bewertung von Innovationsprojekten Souder, W. E., & Jenssen, S. A. Autonomy as key variable]. Wiesbaden, [Process-related valuation of innovation (1999). Management practices influ- Germany: Deutscher Universitätsverlag. projects]. PhD dissertation. Vallendar, encing new product success and failure Lange, E. C. (1993). Germany: WHU. in the US and Scandinavia: A cross-cul- Abbruchentscheidung bei F&E-Projekten Rossi, S. (2005). Using the insan- tural comparative study. Journal of [Stopping decisions for R&D projects]. ity clause for failed projects. Retrieved Product Innovation Management, 16(2), Wiesbaden, Germany: Deutscher September 7, 2005, from 183–203. Universitätsverlag. www.cio.com.au/index.php.id;16272 Tjosvold, D., Tang, M. M. L., & Leavitt, H. J. (1965). Applied 7491; fp;4;fpid;21 West, M. (2004). Reflexivity for team organization change in industry: Rubenstein, A. H., Chakrabarti, A. innovation in China. The contribution Structural, technological and humanis- K., O’Keefe, R. D., Souder, W. E., & of goal interdependence. Group & tic approaches. In J. G. March (Ed.), Young, H. C. (1976, May). Factors influ- Organization Management, 29(5), Handbook of organizations (pp. encing innovation success at the project 540–559. 1144–1170). Chicago: Rand McNally. level. Research Management, 19, 15–20. Wheelwright, S. C., & Clark, K. B. Lechler, T. (1999). Was leistet das Schlaak, T. M. (1999). Der (1992). Revolutionizing product develop- Promotoren-Modell für das Innovationsgrad als Schlüsselvariable. ment. New York: Free Press. Projektmanagement? [What is the con- Perspektiven für das Management von Witte, E. (1973). Organisation für tribution of the Promotor Model to Produktentwicklungen [The degree of innovationsentscheidungen. Das ptomo- project management?]. In J. innovativeness as a leading indicator. torenmodell [Organization for innova- Hauschildt & H. G. Gemünden (Eds.), Perspectives for the management of tion decisions. The Promotor Model]. Promotoren. Champions der Innovation product development]. Wiesbaden, Goettingen, Germany: Vandenhoeck & (pp. 179–209). Wiesbaden, Germany: Germany: Deutscher Universitätsverlag. Rupprecht.

KLAUS BROCKHOFF, PhD, has held business faculty positions at six universities, including the University of Bonn (Germany), University of Kiel (Germany), University of Lund (Sweden), New Jersey Institute of Technology (United States), and currently, since 1999, WHU—Otto Beisheim Graduate School of Management (Germany). A graduate of the University of Bonn and the University of Münster, and a former research fellow at the University of California, Berkeley, Dr. Brockhoff has documented his research in the areas of technology management, innovation management, business strategy, and business policy in 21 books and more than 270 articles, many of which were published in leading international research journals, such as Management Science, Journal of Product Innovation Management, IEEE Transactions on Engineering Management, and Research Policy. He has consulted for numerous government and business organizations and has received two noteworthy awards, the Max Planck Research Award and Karl Heinz Beckurts Award. Dr. Brockhoff is also an elected member of the Berlin-Brandenburg Academy of Sciences, the European Academy of Arts and Sciences, and the Academy of Technology; he sits on the board of six journals, two corporations, and three foundations. He holds an honorary doctorate from the University of Berne (Switzerland).

36 AUGUST 2006 PROJECT MANAGEMENT JOURNAL THE PERCEIVED VALUE AND POTENTIAL CONTRIBUTION OF PROJECT MANAGEMENT PRACTICES TO PROJECT SUCCESS

CLAUDE BESNER, PMP, University of Quebec, Canada BRIAN HOBBS, PMP, University of Quebec, Canada

Introduction ABSTRACT he organizational value of practicing project management is a central theme comprising much of the field’s current research and debate (Thomas & This paper contributes to the ongoing work T Mullaly, 2005). Such value, however, particularly in terms of return on invest- and debate on the value of project management, accomplishing this through an ment (ROI), is one that researchers and practitioners cannot easily calculate for empirical investigation of practitioner per- every aspect of professional practice. Investigating which practices have the poten- ceptions on the relative value of different tial to enhance project performance—and identifying which are perceived as the project management practices and their most valuable—is an alternative method for gauging this value in day-to-day pro- potential to contribute to improved project fessional practice. Many studies analyzing the most valued practices have focused performance. This investigation is based on a large-scale survey of 753 project man- on investigating one aspect of practice: the use of tools and techniques. The signif- agement practitioners. This paper aims to icance of this one aspect of practice, albeit an important one, is readily observable. answer four questions relating to the value This paper presents and discusses the results of a large-scale survey on project of project management. By identifying the management practices. The results of the first part of the survey were presented at most valued practices, practitioners and the third Project Management Institute (PMI) Research Conference (Besner & organizations can identify their priorities when developing their project manage- Hobbs, 2004). These results showed the extent of tools and techniques use. The ment competencies. This can also guide results presented and discussed in the present paper are based on the survey’s sec- the profession in selecting priorities for ond part, which investigated practitioner perceptions of the potential contribution future development. When choosing prior- of tools and techniques to project success. More precisely, it examined which tools ities to develop and implement, organiza- and techniques possess the greatest potential for improving performance through tions can look to the tools that practitioners identify as most valuable, as more extensive or better use. The measurement of the potential for improvement having the most potential for increased was then integrated in a construct to measure the value of each tool. Identification contribution to project performance, and of the most valued practices can identify priorities for individual practitioners and as presently under-utilized. In order to fully individual firms in the development of their project management competencies. understand the nature of project manage- This finding can guide project management professionals in selecting priorities for ment practices, and the mechanisms through which these create value, future development. The paper aims to answer four questions: researchers must better clarify the distinc- 1. Which set of tools and techniques—and therefore, which practices—do tion between the project phases and proj- professionals consider as having the greatest potential and the least poten- ect processes. These findings can help tial, as possessing the most value and the least value? project management professionals in 2. How does this perceived value vary in different contexts and in relation to selecting priorities for future development. the different phases in the project life cycle? Keywords: project management practices; 3. What priorities should practitioners and organizations set when they are tools and techniques; project success choosing to invest in developing project management practices? 4. What future developments in project management practice and theory do ©2006 by the Project Management Institute Vol. 37, No. 3, 37-48, ISSN 8756-9728/03 these results suggest?

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 37 PMI’s PMBOK® Guide (Project Management Institute, The specific contribution of tools to different contexts is 2004) identifies an extensive set of project management another important part of the relation to value creation. tools and techniques, all of which are generally considered Besner and Hobbs (2004) examined the complex reality of valuable and applicable to most projects most of the time the varying use of different sets of tools in relation to con- (p. vii). This publication does not, however, outline the rel- text and provide detail on the variation in project manage- ative importance of the many tools and techniques in the ment practice by project type. Milosevic and project manager’s toolbox. The PMBOK® Guide states that it Iewwongcharoen (2004) explored the contingent use of is necessary to adapt practice to the particular situation by project management tools and techniques and the affect of choosing which tools and techniques to employ, but does this use on project success. Hargrave and Singley (1998) sur- not provide guidance as to which tools are most valuable in veyed project managers in the United States Army Corps of different contexts (p. vii). These considerations are outside Engineers on the use of the 37 processes and 116 techniques the PMBOK® Guide’s scope. and tools. McMahon and Lane (2001) studied the use of At an operational level, information on the relative tools in the specific in relation to the phases of the project value of tools and techniques and on the variations of this life cycle; they classified the tools by phase to underline the value in different situations can have very practical implica- variation in use throughout the project life cycle. tions. Examining the differences in value of tools and techniques and the variations in different contexts and phases is Research on the Value of Project Management Practice also a way to reflect on professional practice at a higher Over the last 30 years, several noteworthy studies have iden- level. Project management is usually primarily associated tified project success factors. Cooke-Davis (2004) summa- with the planning and the controlling of project execution. rized these and proposed a distinction among three levels of This operational view contrasts with the strategic view of project success: Doing projects right, doing the right proj- project management, as conceptualized in organizational ects, and doing the right projects right, time after time. Most project management (Dinsmore, 1999; Project Management of the literature focuses on doing projects right. But as Institute, 2003). The present investigation on the current Cooke-Davis demonstrated, the practices that are associated practice and the perceived value of project management with success are different at each level. The research on suc- tools and techniques can shed light on both the operational cess factors has shown that the question of what constitutes and the strategic roles of practicing project management. success is complex and multifaceted. Although these studies have identified some significant and consistent results, the The Literature on Project Management Tools factors only partially explain project success. The dynamics Many project management tools are inherently value-orient- leading to project success remain largely undisclosed. ed. The practice of value analysis (VA) is devoted to minimiz- Demonstrating the business value of practicing project ing the cost and optimizing the performance of projects and management is one of the major issues in project manage- deliverables. Earned value management (EVM) uses value as ment today. Because of this, it is a high-priority concern for a metric for gauging cost and schedule performance during PMI’s Research Department and the subject of several project implementation. Financial measurement tools—such research efforts recently initiated by PMI (Hobbs, Thuillier, as cost/benefits analysis (CBA)—are also used to measure & Aubry, 2005; Thomas & Mullaly, 2005). But attempts to organizational value. These tools provide useful information find a simple and direct relationship between project man- for implementing rational decision-making processes. agement practice and ROI have failed to find a statistically Besides these value-oriented tools, there are other tools significant link (Ibbs, Reginato, & Kwak, 2004). Such a fail- in the practitioner’s toolbox that have the potential to ure, however, may have resulted from an insufficient sample improve projects’ success and contribute to value creation. size. Researchers have argued that the benefits of project For example, Raz and Michael (2001) examined the use of management practice are not all captured by ROI metrics; risk management tools in Israeli high-tech industries inves- because of this, the field may underestimate the discipline’s tigating the frequency of use, the perceived contribution of impact on innovation (Turner & Keegan, 2004), on process use to project success, and the extent to which use was asso- improvements (Winch, 2004), and on personnel ciated with high performance. Thamhain (1998) studied the (Thamhain, 2004). It is also possible that past research has use and the perceived value of 29 project management tools failed to identify the factors that truly determine project suc- and techniques. He concluded that the contribution of proj- cess. Although this issue is certainly complex, current ect management tools and techniques to project perform- research efforts should help clarify this issue. ance is conditional: Contribution is based on the way The present paper aims to contribute to the study of the project managers integrate these into the project manage- value of project management practice. Successful projects ment process and the way project teams accept these provide value to organizations; project management prac- processes. White and Fortune (2002) examined tool-and- tices provide organizations with a strategic and valuable technique use in relation to project outcomes and project asset. Value is created when good project management prac- success. Their study brought to light many details concern- tices and good measurement tools improve project success. ing the varying levels of usage of project management tools Studying tools and techniques is a tangible way to research and techniques. project management practices because tools and techniques

38 AUGUST 2006 PROJECT MANAGEMENT JOURNAL are directly related to the things practitioners do. These are and the process group. Furthermore, the treatment by the means through which project managers execute project process group rather than by phase begs the question as to management processes. These are also the means project whether there are significant differences among the phases. managers can use to measure dimensions of project per- This study addresses this issue. formance and success: cost, time, quality, progress, satisfac- The front-end of the project has received less attention tion, and other dimensions of success. in the project management literature than the subsequent A project manager’s practical know-how—those skills phases that deal with detailed planning and execution. used to execute processes and practices—is an important Wideman (2002) reviewed the literature on the project life part of the organization’s tacit knowledge asset (Koskinen, cycle and its importance. Morris (1998) argued that “The Pihlanto, & Vanharanta, 2003; Nonaka, 1994). The opera- decisions made at the early definition stages set the strategic tional complexity associated with the integrated use of a framework within which the project will subsequently specific set of tools and techniques represents an intricate develop. Get it wrong here, and the project will be wrong for subsystem of tacit knowledge that is hard to replicate. a long time” (p. 5). The role of the initiation phase in defin- Therefore, organizations can consider the underlying practi- ing the project—and its influence on project success or proj- cal knowledge associated with a set of tools as a strategic ect failure—poses a strong argument for integrating the asset. In order to implement its strategies, organizations initiation phase into the project management domain. must possess the capabilities needed to execute projects. In many organizational contexts, however, a project Jugdev and Thomas (2002) found that these “capabilities only becomes a project after it has been authorized for exe- are combinations of proprietary resources, knowledge, and cution, which takes place after the front-end phase has been skills that become institutionalized into operating routines completed. For example, the front-end often takes place in a and tacit knowledge” (p. 281). customer organization before a request for proposals is Organizations and their project managers must choose made. In this case, the customer front-end is not part of the the sets of tools that comprise their toolbox. They must inte- mandate to the supplier’s project management team. In the grate these tools to practice project management as a means case of in-house projects, non-project personnel often do for building a strategic asset. They should align these tools the front-end. And project management personnel are typi- with the project context. Milosevic and Ozbay (2001) found cally given the project mandate only after it has been that when organizations use a set of context-compatible approved. Thus, the front-end is not part of the project man- project management tools, they enhance their project deliv- agement personnel’s mandate. The project management lit- ery capability. Milosevic (2003) proposed a model in which erature in general—and the PMBOK® Guide (Project an organization’s project management toolbox stands as the Management Institute, 2004) in particular—downplay the foundation for its strategic project management process. In importance of the initiation phase. PMI’s argument for this model, organizations align their toolbox and their doing so is that most project personnel are not involved in choice of project management tools and techniques with this phase (p. vii & p. 43). Data from the present study is the organization’s strategy and with the project environment analyzed to address this issue. and context. The study of the value of project management tools can thus contribute both to immediate practical con- Methodology cerns of tool selection and to higher-level concerns of the A description of project management practice has been built organizational value of project management practice. based on a survey focused on tools and techniques that are specific to project management. In contrast with previous Variations Throughout the Life Cycle research, general concepts and processes (e.g., training pro- The project life cycle can be defined as a sequence of major grams, performance measurement) have been excluded from phases through which the project evolves from beginning to the study. The tools and techniques selected are more specific end, a sequence in which each phase is separated by and closer to day-to-day practice, closer to the things people approval gates. The practice of managing by phase has occu- regularly do. Although this involves a partial view of project pied a prominent position in the project management liter- management practice, it restricts the investigation to those ature and practice for a very long time. The phase in which well-know tools and techniques that are specific to project the project stands at any moment in its life is an important management. Doing so ensured that the practitioners partici- part of its context. The PMBOK® Guide (Project Management pating in the study easily understood the questionnaire. Institute, 2004), however, does not identify management- Figure 1 lists the 70 tools and techniques that were by-phase as a fundamental project management process The included in the survey questionnaire. Use levels vary con- PMBOK® Guide (Project Management Institute, 2004) does siderably, from 1.4 to 4.1, based on a scale ranging from 1 introduce the concept of process group and the idea that (not used) to 5 (very extensive use). Figure 1 shows decreasing processes from these groups are repeated during each phase. levels from left to right and from top to bottom. The process groups of initiating, planning, executing, and Many analyses were performed on the different data closing have names and definitions that are very close to subsets, but the lists for most often used tools and least those used to identify project phases. It is, therefore, not often used tools produced results very similar to most of the always easy to maintain the distinction between the phase subsamples. Thus, the basic toolbox is more or less the same

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 39 for everybody. But there are significant differences between The respondents had the following demographics: specific groups of users. This allowed us to identify distinct • Age: 30–50 (74%) sets of specialized tools. A previous paper (Besner & Hobbs, • Gender: Male (67%); Female (33%) 2004) discussed the data on the use of tools and techniques. • Current primary role: The questionnaire also collected contextual data on – Team member (8%) respondents (position, education, experience, etc.), their – Project manager (51%) organizations (size, industry, project management maturity, – Program manager/director (24%) etc.), and their projects (more than 10 variables). This infor- – Other (17%) mation allows for segmentation of the data to determine how project management practices varied among the differ- More than half (58%) are currently working on projects ent respondents, organizations, and project contexts. The in information technology and telecommunications. This fact that the sample is split evenly for many of these vari- percentage is approximately 5% higher than in PMI mem- ables renders the analysis easier and more reliable. bership. About 12% of the participants reported working on The Web-based questionnaire was completed by 753 engineering and construction projects and another 12% experienced project practitioners, most of whom were PMPs. reported working on business services projects. The respon-

From Limited to Extensive Use From Very Limited to Limited Use Less Than Very Limited Use

Progress report Contingency plans Life cycle cost (LCC) Kick-off meeting Re-baselining Database of contractual commitment PM software for task scheduling Cost/benefit analysis data Gantt chart Critical path method and analysis Probabilistic duration estimate (PERT) Scope statement Bottom-up estimating Quality function deployment Milestone planning Team member performance appraisal Value analysis Change request Team-building event Database of risks Requirements analysis Work authorization Trend chart or S-curve Work breakdown structure Self-directed work teams Control charts Statement of work Ranking of risks Decision tree Activity list Financial measurement tools Cause and effect diagram PM software for monitoring of schedule Quality plan Critical chain method and analysis Lesson learned/post-mortem Bid documents Pareto diagram Baseline plan Feasibility study PM software for simulation Client acceptance form Configuration review Monte-Carlo analysis Quality inspection Stakeholders analysis PM software for resources scheduling PM software for resources leveling Project charter PM software for monitoring of cost Responsibility assignment matrix Network diagram Customer satisfaction surveys Project communication room (war room) Communication plan Project Web site Top-down estimating Bid/seller evaluation Risk management documents Database of historical data PM software multiproject scheduling/leveling Earned value PM software for cost estimating Database for cost estimating Database of lessons learned Product breakdown structure Bidders conferences Learning curve Parametric estimating Graphic presentation of risk information

Figure 1: The 70 tools in decreasing order of average use

40 AUGUST 2006 PROJECT MANAGEMENT JOURNAL dents were specifically asked to indicate the phase(s) of greatest potential for increased contribution to project per- projects during which they are most often involved. Many formance, this is indeed the case. There are seven tools that respondents indicated involvement in more than one phase. appear in both the list of the most often used tools and the Initiation/Concept 52% list of the tools with the greatest potential to contribute to Planning/Development 83% improved project performance. Execution/Implementation 77% • Lessons learned/post-mortems Finalization/Commissioning/Handover 54% • Requirements analysis • Scope statement Statistical significance reported in this paper is from the • Work breakdown structure (WBS) results of t-tests used to verify differences between means • Project management software for monitoring of schedule and chi-square for contextual differences. More complete • Project management software for task scheduling information on the survey and methodology can be found • Project management software for resource scheduling. in Besner and Hobbs (2004). It is also worthy to note that six of the eight proj- The Potential Contribution to Improved Project Performance ect management software functions proposed in the The survey questionnaire makes a distinction between the questionnaire are listed in the top 20 tools with the usefulness of present practices and the potential impact of greatest potential for increased contributions to proj- improved practice on project performance. The left-hand ect performance. column of Figure 2 presents a summary of the results for the There are two potential explanations for this phenome- potential impact of improved practice. non of highly used tools having significant potential for There are four databases among the tools with the increased contribution to performance. The unexploited greatest potential to improve project performance. These potential may involve the possibility of increasing use or of four databases comprise lessons learned, historical data, better use. One possible explanation is that some tools are risks, and cost estimating data. Besner and Hobbs (2004) used often enough but not well enough. This is the case for showed that database tools have low use rates and that this lessons learned, which are often accumulated without fur- low use rate seems related to the project manager’s need for ther application for guiding future projects. It is difficult to organizational support. It is very difficult for individual imagine more frequent use of scope statements than what practitioners to create and use such databases without orga- this study showed. The potential may well involve better— nizational support. not more frequent—use. The first three tools in this list are related to organiza- An examination of the tools with the least potential for tional learning and memory: database of lessons learned, les- increased contribution to project performance also yielded sons learned/post mortem, and database of historical data. some interesting results. An examination of the list of the The databases for lessons learned and historical data have tools with the least potential reveals two types of tools: very limited current use; but practitioners considered these Tools with low use levels and tools with moderate use lev- the tools with the greatest potential to increase project suc- els. In both cases, the respondents reported that their pres- cess rates. Lessons learned/post-mortems are already among ent use enabled them to adequately complete their projects. the most extensively used tools but still have the potential for For tools such as the three tools associated with contractual contributing significantly to improved performance. bidding and the war room, the results indicate that present The use of the concept of the learning organization has use levels are moderate and satisfactory. become widespread in management. Sense and Antoni (2003) Many of the other tools identified as possessing the established a useful distinction about learning from projects: A least potential contribution were also among the least used. lesson learned can be about learning between projects or with- Monte-Carlo analysis is at the very bottom of the list. The in a project. The databases mentioned here are potentially part practitioners surveyed did not value such tools. One could of the organizational infrastructures identified by Sense and argue that the cause for this very poor perception is igno- Antoni, as those resulting from learning between projects. The rance, but the data suggests otherwise. The respondents were post-mortem—during which lessons learned are established— invited to indicate when they had insufficient knowledge of is most often completed at a project’s end; it is potentially a the tool or technique, when they were unable to offer an means for learning between projects. opinion about more extensive or better use. The tools and The list of tools with the greatest unexploited potential techniques identified in this survey are all very well known. contains four tools related to risk management: risk man- The survey results indicate that the respondents were famil- agement documents, ranking of risks, database of risks, and iar with these tools. contingency plans. Practitioners responding to this survey indicated that there is much potential for increasing project The Intrinsic Value of Tools performance through more or better use of risk manage- A variable was developed to measure the intrinsic value of ment tools and techniques. tools, as perceived by respondents. This variable was created Although it is a little surprising to see tools that already by adding the present extent of use to the potential contri- have high use levels appear in the list of the tools with the bution to project performance of more or better use. This

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 41 Potential Intrinsic Value

1 Database of lessons learned PM software for task scheduling 2 Lesson learned/post-mortem Progress report 3 Database of historical data Scope statement 4 Risk management documents Requirements analysis 5 Requirements analysis Kick-off meeting 6 Ranking of risks Gantt chart 7 Database of risks Lesson learned/post-mortem 8 Scope statement Change request 9 Database for cost estimating PM software monitoring schedule 10 PM software monitoring schedule Work breakdown structure 11 Work breakdown structure Milestone planning 12 PM software for multiproject Statement of work 13 Contingency plans PM software resources scheduling 14 PM software resources scheduling Risk management documents 15 PM software for task scheduling Activity list 16 Team-building event Quality inspection 17 PM software for monitoring cost Baseline plan 18 Stakeholders analysis Contingency plans 19 Communication plan Ranking of risks 20 PM software for cost estimating Client acceptance form

• • • • • • • • • 56 Top-down estimating Life cycle cost (LCC) 57 Self-directed work teams Graphic of risk information 58 Learning curve Parametric estimating 59 Work authorization Learning curve 60 Trend chart or S-curve Quality function deployment 61 Network diagram Value analysis 62 PERT analysis Trend chart or S-curve

Lowest Highest 63 Control charts Critical chain method and analysis 64 Bid documents Control charts 65 Bid/seller evaluation PERT analysis 66 Decision tree Cause-and-effect diagram 67 Cause-and-effect diagram PM software for simulation 68 Pareto diagram Pareto diagram 69 Bidders conferences Decision tree 70 Monte-Carlo analysis Monte-Carlo analysis

Figure 2: Unexploited potential and intrinsic value in decreasing order

42 AUGUST 2006 PROJECT MANAGEMENT JOURNAL yielded a measure of the tool’s overall potential to con- Caution should be exercised in interpreting these previ- tribute to project success or its intrinsic value. The result of this ously-mentioned lists because the cut-off in Figure 2 and the measure is presented in the right-hand column of Figure 2. It selection of just the four tools for presentation in each is expressed as follows: group were selected arbitrarily. Present extent of use + Potential improvement = Intrinsic value Adequately Utilized Tools From an examination of Figure 2 and the lists of tools From the previous discussion, one can see that some tools with the highest and lowest intrinsic values come two cate- are important to practitioners and the present use is ade- gories of tools: super tools and discredited tools. quate. In other words, some tools are creating considerable value at their present level of use but increased use is Super Tools neither necessary nor desirable. The following tools Tools with high intrinsic value could be called super tools. showed a higher than average level of use and a lower These are divisible into two groups. The first group con- than average level of potential for contributing to tains the most extensively used and those with the greatest improved performance: potential for increased contribution to project perform- • Activity list ance. These, therefore, score very high on value. Despite • Gantt chart extensive use, these tools still have the potential of con- • Work authorization tributing to increased performance if more or better use is • Self-directed work teams made of these. Their high value is attributable to the com- • Top-down estimating bination of these two factors. The following are this group’s • Bid documents four most valued tools: • Client acceptance form. • Software for task scheduling • Scope statement The first three tools are among the most extensively • Requirements analysis used. The others are in the middle range with respect to • Lessons learned/post-mortem. actual application. All are well understood and present use was reported as satisfactory. Organizations already using Another group of super tools also shows very high these tools should probably continue doing so. Others not scores for use, but does not show high scores for potential using these tools regularly might consider adopting these. improvement. Theses tools are very valuable but are usually used at levels close to their full potential. The following are Underutilized Tools this group’s four most valued tools: From the previous discussion, one can see that some tools • Progress report possessing a considerable potential to contribute to improved • Kick-off meeting performance are underused. The following tools presently • Gantt chart show a higher than average level of potential and a lower than • Change request. average level of use: • Database of lessons learned Discredited Tools • Database of historical data Most of the tools with the least intrinsic value are tools that • Database of risks are rarely used and are perceived as having very little poten- • Database for cost estimating tial. Next are this group’s four least valued tools: • Database or spreadsheet of contractual commitment data • Monte-Carlo • Project management software for multiproject • Decision tree analysis scheduling/leveling • Pareto diagram • Project management software for monitoring of cost • Cause and effect diagram. • Project management software for cost estimating • Project management software for resources leveling This evaluation underscores the need to reconsider the • Earned value position of these tools in the project management literature • Feasibility study and training as well as in the PMBOK® Guide (Project • Stakeholders analysis Management Institute, 2004). However, practitioners con- • Configuration review sider some tools with very low intrinsic value to have some • Graphic presentation of risk information. potential, even if they infrequently use these tools, these include the following: Organizations can consider these underutilized tools as • Project management software for simulation potential investment and development opportunities. The • Critical chain method and analysis list contains five different types of databases. To implement • Value analysis and use these tools, project managers would require organi- • Quality function deployment. zational commitment and support. The survey results suggested that such investments are worth considering because

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 43 the practitioners believe that these tools will contribute to improved project performance. 90 As shown in Figure 1, project managers already use three project management software tools extensively. These 80 tools are all related to scheduling. The survey participants identified four additional project management software tools that are underutilized; these involve relatively complex 70 or sophisticated application of project management software with significant potential to contribute to improved 60 performance. These underutilized tools include earned value, stakeholder analysis, and feasibility study. A word search in the PMBOK® Guide (Project 50 % of Respondents Involved Management Institute, 2004) reveals several references to Program Manager/Director Project Manager these underutilized tools. PMI even has the College of 40 Performance Management that promotes the use of earned Team Member Total value; it also publishes a standard on the subject. Identifying earned value and stakeholder analysis as important and 30 underutilized tools validates the already significant status of Initiation Planning Execution Finalization these tools within PMI standards. The PMBOK® Guide’s ref- Generic Project Life Cycle Phases erences to feasibility studies place it outside the scope of project management. Identifying the feasibility study as an Figure 3: Involvement of different project management roles in important but underutilized tool reinforces the need to different project phases increase its importance in the project management literature and practice. Moreover, the feasibility study is related to the most contexts. At the same time, systematic significant dif- strategic front-end phase, as discussed in the next section. ferences exist. For example, almost all tools and techniques The list of underutilized tools with significant potential also are more valued by practitioners working on large projects includes the configuration review and the graphic represen- and in organizations with high levels of project manage- tation of risks. ment maturity. About half of the tools are more valued for external projects and for long-duration projects. Practically Setting Priorities for Development and Implementation none are significantly more valued in the opposite contexts The most valued tools and the most underutilized tools can (small projects, low maturity organizations, internal or provide organizations with the guidance they need in devel- short duration projects). A discussion of one important oping and implementing project management tool and aspect of context, the generic phases of the project life cycle, techniques. The survey respondents identified the tools and focusing on the initiation phase follows. techniques that organizations should develop and implement to improve project performance. Inversely, partici- Involvement in Different Phases pants believe that the least valuable tools and the tools with The survey questionnaire reports data on the respondents’ the least potential are poor investment choices. The project management positions and their involvement in dif- PMBOK® Guide states that organizations must adapt their ferent phases of the project life cycle. This information is choice of appropriate tools and techniques to match their presented in Figure 3. specific projects and contexts (2004, p. 3). Next, this paper As one would expect, participant involvement is highest in address the variations involved when practicing project the planning and execution phases. However, practitioners management in different contexts, focusing on the differ- demonstrated significant involvement in both the initiation ences among the project life cycle’s phases. (52% of respondents) and closing phases (54% of respondents). Although participant involvement in the initiation Variations in the Value of Tools and Techniques in Different Contexts phase is high for the entire sample, it is the program man- Besner and Hobbs (2004) showed that the basic project agers/directors who are particularly active during this phase. management toolbox is very similar across different con- It is during this phase when organizations align the project texts. The common pattern that exists across the project with their needs and strategy. The high percentage of senior management community constitutes the generic pattern of personnel involvement reflects this. Further analysis showed practice that is applicable to almost all projects in almost all that program and project managers do not use different sets contexts. This generic practice is the basis of the PMBOK® of tools during the initiation phase, indicating that there is Guide (Project Management Institute, 2004). The authors a similarity in the nature of these roles during initiation. also found significant and important differences in relation An analysis of the socio-demographic data revealed that to working in a different context. The same is true of the per- the respondents and the characteristics of their organiza- ceived value of tools and techniques in different contexts. tions and projects showed no significant relationships The set of most valued tools (Figure 2) is consistent across between their involvement in the different phases, their per-

44 AUGUST 2006 PROJECT MANAGEMENT JOURNAL sonnel characteristics (sex, age, or education level), the pro- The greater value of these tools in a particular phase ject’s size or complexity, their organizational maturity level does not mean that these are not valued during the other or overall size—or the fact that the projects have internal or phases. Four of these tools are included in both the list of external customers. the ten most used and the list of the ten most valued tools. There is a significant relationship, however, between Despite the generally recognized value in the overall sample, respondents’ involvement in the initiation phase, their hier- the four tools in the following list show statistically signifi- archic level (as measured by level of authority (p < 0.000), cant variations in value by phase: and their project role, as shown in Figure 3 (p < 0.000). • Project management software for task scheduling There is also a significant relationship between one’s • Scope statement involvement in the initiation phase and the presence of • Requirements analysis both a multiproject environment (p = 0.003) and multidis- • Work breakdown structure. ciplinary teams (p = 0.003). Further analysis revealed that one’s involvement in this phase is associated with the busi- The kick-off meeting is among the five most used and ness development function (p < 0.000), with formal train- valued tools, but it did not show statistically significant ing in business (p = 0.006), and less strongly, with variation in value by phase. As a result, it is not listed in involvement in business services projects (p = 0.038). The Figure 4. The kick-off meeting is clearly associated with relationship with business development is natural given the initiation process; it is repeatable for each new set of that business development takes place during the early activities throughout the project life cycle. The initial front-end of project initiation. The relationship with formal project kick-off often plays a very important role during education in business is indicative of the skill set required the initiation phase. However, some aspects of the initia- in project initiation. tion phase are specific to this phase and are not typical of the initiation process in other phases. It is easy to under- Most Valued Tools by Phase stand why cost/benefit analysis and feasibility studies are The comparisons between most and least valued tools for important during the initiation phase. It is not, however, groups of respondents participating in different phases pro- easy to see these as important tools during other phases. duce almost identical lists to those shown in Figure 2. This is The most obvious observation from Figure 4 is that in part due to the fact that most of those reporting involve- the initiation phase is very different from the other phas- ment in the initiation phase also report involvement in other es. The activities of this phase are quite specific. To say phases. It is also because even in the initiation phase, the typ- that the initiation phase and the initiation processes in ical processes of planning and control are applied to the each phase are the same is to underestimate these differ- phase’s specific activity. Nevertheless, statistically significant ences. The use of the same term—initiation—for both the differences were also revealed. Figure 4 shows the tools that phase and the process, as is the case in the PMBOK® are significantly more valued in each phase; the plus sign in Guide (Project Management Institute, 2004) underesti- the columns shows that those tools are significantly more mates the specific nature of the initiation phase and can valued by those participating in the specified phase as com- generate confusion. pared to those not participating in this phase (p<_0.01). The list’s first three tools are directly related to choosing the best project or finding the best solution to Tools Value Init. Plan. Exec. Final. the project mission. These refer to the strategic role of the front-end phase of the project. The feasibility study has Cost/benefit analysis + – – been identified above as an underutilized tool. Because Feasibility study + the PMBOK® Guide (Project Management Institute, 2004) Financial measurement tools + excludes project initiation from the scope of most proj- Scope statement + ects, it is not surprising that this publication does not Work breakdown structure + highlight cost-benefit analysis and feasibility studies. PM software for cost estimating + The set of tools identified as being valued during the Responsibility assignment + initiation phase appears very well integrated. The general matrix scope of the project is first determined during the initia- PM software for resources + tion phase. The first scope statement and the correspon- scheduling ding higher levels of the WBS—for which responsibility is PM software for resources + then assigned to key project resources—are crucial output leveling decisions made during the front-end phase. The responsi- Stakeholders analysis + + bility assignment matrix is a structure that relates the proj- Requirements analysis + + + ect organization structure (more specifically, all project Team-building event + stakeholders) to the WBS. This ensures that responsibili- PM software for task scheduling + ty is assigned for each element of the project’s scope of work. The responsibility assignment matrix can be direct- Figure 4: Tools showing significant differences in values in each phase ly linked to the theory of management-as-organizing, as

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 45 opposed to management-as-planning, as discussed by planning and development phase, the requirements are ana- Koskela and Howell (2002). According to Koskela and lyzed to clarify their meaning, to elaborate upon—and to Howell, managing-as-organizing helps bring together man- develop—detailed technical specifications, and to plan the agement and action, which are often disconnected, in the tasks necessary to meet requirements. During termination, management-as-planning view. The higher level of authori- the deliverables are analyzed to determine if the require- ty of the practitioners participating in this phase confirms ments have been met. the phase’s more strategic nature. Requirement analysis is one of the “super tools” previ- Practitioners participating in the initiation phase apparent- ously identified. It is among the most used and the most val- ly recognize that the major goals of the strategic front-end ued tools. The very definition of project management is phase are planning the right allocation of resources and finding centered on meeting requirements: “The application of the right people to manage the key deliverables identified in the knowledge, skills, tools and techniques to project activities WBS. Resource scheduling and leveling in this phase are, there- to meet project requirements” (Project Management fore, related to the “rough-cut-capacity-planning” as described Institute, 2004, pp. 8 & 368). A word search in The PMBOK® by Hendricks, Voeten, and Kroep (1999). As previously men- Guide reveals that the expression “requirement analysis” is tioned, respondents’ involvement in the initiation phase is sig- not used in the Guide. Rather, the PMBOK® Guide identifies nificantly related to their presence on multidisciplinary teams requirements as one of the important elements of the proj- and in multiproject environments. In this context, long- or ect charter that is issued by the project sponsor, by the proj- medium-term resource allocation is an important function ect initiator that operates outside the project organization of program and portfolio management. Cost concerns sug- (pp. 81–82). The PMBOK® Guide identifies requirements as gested by attributing greater value to project management existing prior to and outside of the project: It does not iden- software for cost estimating are related to allocating tify these as the object of analysis. This survey’s results indi- resources and to evaluating the project’s cost and benefits, cate that requirement analysis is a very important activity for possibly in terms of ROI. project practitioners to perform. This evidence demonstrates Requirements and stakeholder analysis are highly val- that requirements analysis is within the scope of the project ued during both the initiation and the finalization phases. and that PMI should include it within future versions of the During project initiation, identifying requirements is very PMBOK® Guide. closely related to identifying stakeholder expectations. Because two key and related closeout activities involve veri- Conclusions fying that the project meets its requirements and the stake- Setting Priorities for Development and Implementation holder expectations, it is not surprising to see that these two Both individual organizations and practitioners—and the tools are valuable elements during both closeout and initia- field of project management as a whole—can identify tion. Requirements analysis is also highly valued during the ways to develop and enhance their project management planning and development phase. The work on require- practices by examining the tools identified in this study as ments during the planning and development phase is more most valuable, as having the most potential for increased focused on technical elaboration and is less tightly related to contribution to project performance, and as presently stakeholder expectations. under-utilized. For example, this study identified those An element often related to commissioning, handover, tools related to organizational learning and memory as implementation, and ramp-up of operations is project termi- among the tools showing the greatest potential for nation. At commissioning, a new group of stakeholders gets improving project performance. The results of this survey involved in the project. These individuals will take charge of also indicate that the current set of well-known project project deliverables. This may also explain the importance of management tools and techniques is more highly valued stakeholder analysis at this stage of the project. in the context of large projects for external customers and less highly valued for smaller projects for internal cus- A Further Examination of “Requirement Analysis” tomers. Given the very large number of these latter types Koskela and Howell (2002) challenged the traditional theo- of projects, the field should focus its efforts on developing retical view of project management. They proposed the a new set of project management tools and techniques, “value generation” view as part of a new enlarged theory of one that focuses on small and internal projects. The devel- project management that includes the fundamental aspect opment of a project management tool set for a specific of customer requirements and therefore of business pro- organization will, of course, need to be based on an analy- pose. The findings from the present study provide detailed sis of the current state of practice in the organization and empirical evidence supporting Koskela and Howell’s the specific characteristics of the projects being managed assumptions. and the organizational context. Requirement analysis may be used to different ends in the different phases. The list of requirements is an important The Specific Characteristics of Project Initiation output of the initiation phase. The use of requirement analy- One purpose of this paper has been to investigate project sis during initiation focuses attention on the production practice through each phase. Slightly more than half of and validation of the project’s requirements. During the the survey respondents reported substantial involvement

46 AUGUST 2006 PROJECT MANAGEMENT JOURNAL in the initiation phase. This brings into question the argu- Ibbs, C. W., Reginato, J. M., & Kwak, Y. H. (2004). ment that this phase is outside the scope of most practi- Developing project management capability: Benchmarking, tioners’ project work. If the initiation phase is as critical as maturity, modeling, gap analysis, and ROI studies. In P. W. some of the literature claims, then downplaying this G. Morris & J. K. Pinto (Eds.), The Wiley guide to managing phase reduces the emphasis on a subject that is critical to projects (pp. 1214–1233). Hoboken, NJ: John Wiley & Sons. project success and value creation. Jugdev, K., & Thomas, J. (2002). Blueprint for value cre- The analysis has shown that the front-end phase has ation: Developing and sustaining a project management some very specific characteristics. It draws upon business competitive advantage through the resource based view. skills to make greater use of tools directly associated Proceedings of the PMI Research Conference 2002. Seattle, WA. with the strategic front-end, such as choosing the right Koskinen, K. U., Pihlanto, P., & Vanharanta, H. (2003). project and managing the organizational interfaces— Tacit knowledge acquisition and sharing in a project work stakeholder analysis, cost/benefit analysis, feasibility context. International Journal of Project Management, 21, studies, responsibility matrix, and resources oriented 281–290. tools, among others. Koskella, L., & Howell, G. (2002). The underlying theo- The authors conclude that the initiation phase is ry of project management is obsolete. Proceedings of the PMI important and specific: In order to adequately portray Research Conference 2002, Seattle, WA, USA. project initiation, both the initiation phase and the initi- McMahon P., & Lane, J. D. (2001). Quality ation processes occurring during each project phase must tools/techniques and the project manager. Proceedings of be taken into account. The distinction between the two the 33rd Annual PMI Seminars and Symposium, Nashville, needs to be made explicit and to be used consistently, oth- TN, USA. erwise confusion is likely. Downplaying one at the expense Milosevic, D. Z. (2003). Project management toolbox. of the other leads to an incomplete view of project man- Hoboken, NJ: John Wiley & Sons. agement practice. Milosevic, D. Z., & Iewwongcharoen, B. (2004). Management tools and techniques: The contingency use Acknowledgment and their impacts on project success. Proceedings of the The authors wish to thank Dr. Lynn Crawford for her gen- PMI Research Conference 2004, London, UK. erosity in allowing them to use questions on contextual vari- Milosevic, D. Z., & Ozbay A. (2001). Delivering proj- ables drawn from her research on project management ects: What the winners do. Proceedings of the 33rd Annual competency. The authors also wish to thank Harry Stefanou, PMI Seminars and Symposium, Nashville, TN, USA. previous Manager of PMI’s Research Department, for his Morris, P.W.G. (1998). Key Issues in Project support in data collection. Management. In J. K. Pinto (Ed.), Project Management Institute Project management handbook. New York, John Wiley. References Nonaka, I. (1994). A dynamic theory of organization- Besner, C., & Hobbs, J. B. (2004). An empirical inves- al knowledge creation. Organization Science, 5(1), 14–37. tigation of project management practice: In reality, which Project Management Institute. (2003). OPM3® tools do practitioners use and value? In D. Slevin, D. Knowledge Foundation. Newtown Square, PA: Project Cleland, & J. Pinto (Eds.), Innovations: Project management Management Institute. research 2004 (pp. 337–351). Newtown Square, PA: Project Project Management Institute. (2004). A guide to the Management Institute. project management body of knowledge (PMBOK® guide). Cooke-Davis, T. (2004). Project success. In P. W. G. Newtown Square, PA: Project Management Institute. Morris & J. K. Pinto (Eds.), The Wiley guide to managing Raz, T., & Michael, E. (2001). Use and benefits of tools projects (pp. 99–121). Hoboken, NJ: John Wiley & Sons. for project risk management. International Journal of Project Dinsmore, P. C. (1999). Winning in business with enter- Management, 19, 9–17. prise project management. New York: AMACOM. Sense, A. J., & Antoni, M. (2003). Exploring the politics Hobbs, B., Thuillier, D., & Aubry, M. (2005). Modeling of project learning. International Journal of Project Organizational Project Management and PMO Performance. Management, 21, 487–494. Research proposal presented to the Project Management Thamhain, H. J. (1998). Integrating project manage- Institute, Research Department. ment tools with the project team. Proceedings of the 29th Hargrave, B. L., & Singley, J. (1998). PMBOK: A guide Annual PMI Seminars and Symposium, Long Beach, CA. for project management in the next century. Proceedings of Thamhain, H. J. (2004). Linkages of project environ- the 29th Annual PMI Seminars and Symposium, Long Beach, ment to performance: Lessons for team leadership. CA, USA. International Journal of Project Management, 22(7), 533–544. Hendricks, M. H. A., Voeten, B., & Kroep, L. (1999). Thomas, J., & Mullaly, M. E. (2005). Understanding the Human resource allocation in a multi-project R&D envi- value of project management. Retrieved August 10, 2005, from ronment. International Journal of Project Management, www.pmi.org/prod/groups/public/documents/info/pp_sp 17(3), 181–188. onsoredprojects.asp

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 47 Turner, R. J., & Keegan, A. E. (2004). Managing technol- Wideman, M. (2002). The role of the project life cycle (life ogy: Innovation, learning, and maturity. In P. W. G. Morris span) in project management. Retrieved May 10, 2004, from & J. K. Pinto (Eds.), The Wiley guide to managing projects (pp. http://www.maxwideman.com/papers/plc- 576–590). Hoboken, NJ: John Wiley & Sons. models/intro.htm White, D., & Fortune, J. (2002). Current practice in proj- Winch, G. H. (2004). Rethinking project management: ect management—An empirical study. International Journal of Project organizations as information processing systems? Project Management, 20(1), 1–11. Proceedings of the PMI Research Conference 2004, London.

CLAUDE BESNER, PMP, holds a degree in architecture, an MBA, and a PhD in management. He is a professor in the Management and Technology Department at the University of Quebec at Montreal (UQAM). He is a past director of the UQAM master’s program in project management, which is accredited by PMI’s Global Accreditation Center. Dr. Besner has more than 15 years of experience as consultant and project director. He is active internationally in the project management community and has presented papers at both research and professional conferences organized by project management organizations worldwide.

BRIAN HOBBS, PMP, holds a degree in industrial engineering, an MBA, and a PhD in management. He has been a professor at the University of Quebec at Montreal in the master’s program in project management for more than 20 years. This program, of which he is a past director, is accredited by PMI’s Global Accreditation Center. Dr. Hobbs is very active internationally in both the project management professional and research communities. He served a three-year term on PMI’s Standards Members Advisory Group (MAG) ending in 2002. He is a reviewer for both the Project Management Journal and the International Journal of Project Management. He has presented several papers at both research and professional conferences organized by PMI and other project management organizations worldwide. In recent years, he has been the member of three teams of researchers that have been awarded competitively bid research grants by PMI. He will be the chairperson of the UQAM Research Chair in Project Management, which will be created in the fall of 2006.

48 AUGUST 2006 PROJECT MANAGEMENT JOURNAL MECHANISMS FOR INTER-PROJECT INTEGRATION–EMPIRICAL ANALYSIS IN PROGRAM CONTEXT

PERTTU DIETRICH, Helsinki University of Technology BIT Research Centre

Introduction ABSTRACT roject management researchers have recently started to recognize that projects and project management do not only represent operational tools to This exploratory study focuses on the problem P organize minor tasks, but strategically important organizational capabili- of inter-project integration and how the perceived uncertainty and structural complexity ties (Artto & Dietrich, 2004; Cleland, 2002; Gareis, 2004). This ongoing discus- affect the importance of different integration sion on the strategic role of projects has led to an emergence of a novel research mechanisms used. Four case programs from area related to management of multiple simultaneous projects organized as four companies were chosen as a source of multiproject programs. Multiproject programs represent vehicles that are empirical data. Data collection in the selected increasingly used to develop and implement strategic organizational changes, case programs was multifaceted and included in-depth interviews, questionnaires, and docu- too complex or vague in their objectives to fit into the traditional project man- ments and archives. Fifteen integration mecha- agement frame. nisms were identified and further categorized Several authors have presented their concerns that the project management into five different classes based on the formal- literature is too heavily based on the ideology that projects are fundamentally ity of the mechanism and whether the integration mechanism is personalized or similar (Milosevic &Patanakul, 2002; Shenhar, 2001; Shenhar & Dvir, 1996). impersonalized in nature. This study provides Guided by these concerns, some research attempts have focused on factors that valuable insights for both researchers and pro- make projects essentially different from each other. A project’s size, uncertainty, gram managers challenged to introduce strate- complexity, and pace are examples of factors that are suggested to have major gic or large-scale changes in organizations. The study demonstrates that proper mecha- impacts on a project’s execution strategy (Milosevic and Patanakul, 2002; nisms for inter-project integration are depend- Shenhar, 2001; Shenhar and Dvir, 1996, Tatikonda and Rosenthal, 2000). These ent on the uncertainty and complexity of the factors resemble those suggested in classical contingency theory, in which differ- program, and that the proper integration strat- ent modes of organizing are explained by complexity, uncertainty and size egy in multi-project programs should allow the (Donaldson, 2001). Thus, a similar kind of thinking is starting to emerge among use of various alternative mechanisms for information delivery. project management researchers. Researchers studying programs and their management have observed a need Keywords: inter-project integration; project uncertainty; structural complexity; to extend the emerging contingency thinking from single-project context to mul- information delivery tiproject context (Pellegrinelli, 1997; Ribbers & Schoo, 2002; Vereecke, Pandelare,

©2006 by the Project Management Institute Deschoolmeester, & Stewens, 2003). However, the existing literature on program Vol. 37, No. 3, 49-61, ISSN 8756-9728/03 management is still heavily based on the “one-size-fits-all” thinking. Moreover, it is not sufficiently explored to determine whether the results from the studies on contingent single-project management apply to multiproject setting. This study aims to open up the contingency view to program management. To be more exact, this study focuses on the problem of inter-project integration and how the perceived uncertainty and structural complexity affect the perceived importance of different integration mechanisms used. Two research questions guide this study:

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 49 • What kind of different mechanisms are used to ensure the delivery process is independent on delivery processes of effective integration between different projects in intra- other tasks, it is called pooled interdependence. Second, organizational development programs? when the output of one task is an input for a second, it is a • How does the perceived uncertainty and structural question of sequential interdependence. Finally, reciprocal complexity affect the perceived importance of the interdependence means that the output of each task mechanisms? becomes input for others. In addition, Thompson (1967) • In order to answer the research questions, this study argued that when the interdependencies among organiza- reports findings from the empirical analysis of inter- tional tasks grow from pooled to reciprocal, integration project integration mechanisms in four case programs. among tasks through rules and procedures does not suffice anymore, but participatory mechanisms such as mutual Problem of Integration adjustments are needed to deliver new information during The problem of integration is a consequence of organiza- the process of action. tional fragmentation. It is proposed that the external require- Finally, Van de Ven, Delbecq, and Koenig (1976) pro- ments emanating from the environment leads to posed that the use of different integration mechanisms is segmentation of organizations into various units (subsys- contingent on three different factors: perceived task uncer- tems), each having its own responsibilities and tasks tainty, work flow interdependence, and work unit size. Their (Lawrence & Lorsch, 1967). Organizational segmentation study on integration modes in 197 formal work units enables organizational subsystems to focus the attention exposed that integration through impersonal mechanisms, effectively on a particular problem or task. However, in order such as plans and schedules, is negatively correlated with to accomplish the overall purpose of the organization, these perceived task uncertainty and positively correlated with organizational subsystems have to be linked together work unit size. In addition, they found that the use of hori- (Lawrence & Lorsch, 1967; Thompson, 1967). In organiza- zontal communication channels is positively correlated tion theory, the management of the linkages among different with the perceived task uncertainty. Moreover, the use of organizational subsystems is called integration or coordina- scheduled meetings as integration mechanisms was tion (Hage, Aiken, & Marrett, 1971; Van de Ven, Delbecq, & observed to correlate positively with perceived task uncer- Koenig, 1976). tainty and workflow interdependence. Finally, they found Different studies in the organizational field reveal that that perceived task uncertainty correlated positively with the the integration between different tasks or activities often use of unscheduled meetings as an integration mechanism. determines how effectively and efficiently the overall goals The three studies have all contributed to research on are achieved (Gittell, 2002). For example, in product devel- integration by showing that different kinds of conditions opment, the way that the work is broken down and the inte- external and internal to the organization lead to different gration among different tasks has a significant impact on kinds of integration requirements, and that the effectiveness productivity, quality, and development time (Clark & of integration is dependent on the fit between selected inte- Fujimoto, 1991; Cohen & Regan, 1996). Gittell (2002) gration mechanisms and the existing integration needs. noted that similar kinds of results have been achieved in Within organization theory, the studies of integration apparel production (Albernathy, Dunlop, Hammond, & are largely focused on the integration mechanism among Weil, 1999), air travel (Gittell, 2001), and healthcare deliv- different parts of the “permanent organizational arrange- ery (Argote, 1982). ments,” more precisely integration between formal work Three different studies have had major impact on units (Van de Ven et al., 1976) or functional departments increasing our understanding of organizational integra- (Hage et al., 1971; Lawrence & Lorsch, 1967), to mention a tion. First, Lawrence and Lorsch (1967) proposed that the few. Another area of interest has been integration in teams appropriate way for integration among different organiza- or groups (Gittell, 2002; Perlow, Gittell, & Katz, 2004). Even tional units is dependent on the degree of differentiation if the problem of integration has been studied extensively in among the units. Moreover, in their study, Lawrence and different kinds of organizations and in team and group Lorsch found that the level of differentiation is consistent arrangements, relatively little is known about integration in with the diversity of the different parts of the environ- complex multiproject entities—i.e., programs. ment. Thus, they concluded that the integrating mecha- Programs are multiproject entities that differ from the nisms in an effective organization are consistent with the “permanent” organizations through their temporally limit- diversity of the environment so that, the more diverse the ed life, and through their action orientation. Programs are environment, the more differentiated and integrated often characterized by a combination of uncertainty related the effective organization. to goals and tasks, and complexity emanating from large Second, Thompson (1967) suggested that the type of size and numerous dependencies among different activi- interdependence among organizational tasks determines ties. In addition, programs require the involvement of the appropriate integration mechanisms in organizations. many individuals and the integration of knowledge from He categorized interdependencies among different organi- various disciplines, and are constantly subject to influences zational tasks into pooled, sequential, and reciprocal. When and development emanating from the external environment each task delivers a discrete contribution to the whole and (Pellegrinelli, 2002). Consequently, integration requirements

50 AUGUST 2006 PROJECT MANAGEMENT JOURNAL for multiproject programs are different from the integration Finally, Adler (1995) examined integration in 13 inter- requirements of permanent organizations. Furthermore, it is departmental product development projects. The distinct proposed that programs differ fundamentally from single integration mechanisms characteristic of the product devel- projects because they often produce not just a single, clearly opment context that they observed in their study include defined deliverable, but rather multiple deliveries. In addi- compatibility standards, capabilities development sched- tion, single projects are often focused on delivering an asset ules, coordination committees, joint development, design or a change, whereas programs aim to produce strategic or rules, tacit knowledge, producibility design reviews, joint extra-project objectives (Pellegrinelli, 1997). teams, exception resolution plans, and transition teams. The The empirical studies on integration in multiple-project results of the study reveal that the use of integration mecha- contexts are relatively few, if any. However, some empirical nisms is contingent on task analyzability and novelty. They studies on integration in the single-project environment have found that decreasing analyzability in the projects requires been accomplished that might help us to understand some additional integration effort in the later phases of the proj- aspects of integration in the multiproject environment. First, ect, and increasing novelty in the projects requires the use of Andres and Zmud (2001) studied the effects of task interde- more interactive integration mechanisms, such as mutual pendence, integration strategy, and goal conflict on the suc- adjustment and team coordination. cess of software projects through laboratory experiment. The Based on the previous studies of integration in perma- results of the study revealed that organic integration strategy nent organizations (Lawrence & Lorsch, 1967; Thompson, characterized by decentralized structure, informal communi- 1967; Van de Ven et al., 1976) and in single-project context cation and cooperative decision-making provided higher (Adler, 1995; Andres and Zmud, 2001; Kraut & Streeter, productivity than mechanic integration strategy character- 1995; Nidumolu, 1996), it is assumed that uncertainty and ized by centralized structure, formal communication, and complexity serve as principal contingency factors, explain- unilateral decision-making. In addition, organic integration ing the adoption of different integration mechanisms in strategy was found to be especially effective when highly multiproject programs. However, since the multiproject interdependent tasks were included in the project. context differs from the permanent organizations’ context, Second, Kraut and Streeter (1995) studied integration and from single-project context, it is expected that the inte- techniques in software development projects. They defined gration needs and the strategies to respond to those needs five categories for different integration techniques. First, for- will be different in multiproject context. This study aims to mal impersonal integration procedures referred to written extend the emerging contingency thinking from single- requirements documents, modification request tracking and project context to multiproject context by exploring the data dictionaries. Second, formal interpersonal integration effects of complexity and uncertainty on inter-project inte- techniques referred to requirement review meetings, status gration in multiproject programs. The methods for the review meetings, and code inspection meetings. Third, infor- empirical exploration of the phenomenon are explained in mal interpersonal procedures referred to unscheduled group the next section. meetings or co-location of requirements and design staff. Fourth, electronic communication—such as electronic mail Methodology and electronic bulletin board—was classified as one distinct Data Collection integration technique. Finally, interpersonal networks referred Inductive, multiple case-study strategy was selected for this to integration through individuals’ interpersonal contacts study (see Eisenhardt, 1989). Four case programs from four outside the projects. The results of their study revealed that different companies were chosen as a source of empirical the use of formal impersonal, and interpersonal, procedures data. Of the four selected case programs, one was already correlates positively with the size of the project. The study also completed and three were ongoing. Data collection in the showed that informal interpersonal procedures were used selected case programs was multifaceted and included in- especially in the planning stage of the project. In addition, the depth interviews, questionnaires, and documents and results suggest that electronic communication was used more archives. In each case, 6 to 11 informants were interviewed. A often when the project was dependent on other groups in the total of 33 interviews were conducted, of which two were organization. Finally, the use of interpersonal networks corre- open and 31 semi-structured. Notes were taken during all the lated positively with a project’s small size, certainty and interviews. In addition, of the 33 interviews, 30 were tape- dependency of input from other groups. recorded and transcribed. The informants served various Third, Nidumolu (1996) studied the affect of require- roles in the case programs, such as program managers, proj- ment uncertainty and integration mechanisms on the per- ect managers, project employees, and members of the pro- formance of 64 information systems projects. The results of gram steering group. The collection of empirical data was the study revealed that the vertical integration through deci- performed during the three-month period from September sions by authorized entities, such as project managers or steer- to November of 2005. ing committees, enables project teams to reduce project risk In order to measure the perceived uncertainty and effec- and uncertainty, and horizontal integration through mutual tiveness of integration in case programs, the interviews were adjustments and communications correlates with improved complemented with 30 structured questionnaires. The fill- project performance. ing out of questionnaires was integrated into the interviews.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 51 Researchers monitored and tape-recorded the filling out of launch in spring 2007—is used to integrate and unify prac- questionnaires in order to ensure that the respondents tices with several customers of the organization. understood the questions. This process of monitoring also Case Delta is an ongoing strategic organizational devel- gave respondents the opportunity to comment on the ques- opment program in a large, public-sector Finnish organizations and to explain their choices, when necessary. In the tion. The program was established in December 2004 by the questionnaire, the 7-point Likert scale was used. head of the organization in order to develop services provided by, and internal processes of, the organization. The aim of Data Analysis the program is to both increase the quality of the services and The first phase of the analysis of qualitative interview data the intra-organizational processes, and to decrease the unit included in-depth, within-case analysis. This within-case costs related to producing service products for customers. analysis phase resulted in an initial list of different integration mechanisms that was further complemented dur- Analysis of Effectiveness in Inter-Project Integration ing the cross-case analysis phase. Second, in order to In order to study the challenge of inter-project integration in compare different cases the level of perceived uncertainty, case programs, it was essential to first measure how well the structural complexity, and perceived integration effective- integration efforts succeeded in the case programs. The eval- ness was analyzed in each case. The analysis of perceived uation of integration effectiveness was based on the analysis uncertainty and effectiveness of integration was based on of four different indicators: participants’ awareness of the the data acquired through questionnaires. The structural situation of different projects in a program, participants’ complexity was defined based on the analysis of inter- awareness of the linkages between different projects in a views, and documents such as program plans, schedules, program, adequacy of communication among different and monitoring reports. projects in a program, and the integrity of projects’ results. Finally, the cross-case analysis included the compari- In order to measure the values for the indicators, each son of integration mechanisms in different cases. This informant was asked to commit to four statements in a 7- final phase of the analysis resulted in a complete list of dif- point Likert scale. For more in-depth information of aver- ferent integration mechanisms and their perceived impor- aged values for each statement and the respective ranges in tance for each case. Moreover, the cross-case analysis answers, see Table 4 in Appendix 1. provided a suggestion of the effects of perceived uncertainty, The results of the analysis reveal that integration effec- and the structural complexity of the importance of different tiveness in case companies varied from moderate (Alpha integration mechanisms. and Delta) to high level (Beta and Gamma). From an integration perspective, the cases of this study do not represent Case Programs either extremely successful ones or total failures. Overview of the Case Programs Because all the case programs are characterized by either Case Alpha represents a strategic intra-organizational devel- a moderate or high level of averaged integration effective- opment program executed in a medium-sized, private-sector ness, it can be assumed that adopted integration mecha- Finnish organization. The program was initiated in March nisms in each case fit with the integration requirements of 2000 by the CEO of the organization in order to respond to the respective case. However, the fact that the range of opin- the changes in the society’s monetary politics and to ration- ions of integration effectiveness is rather wide, especially in alize the internal information delivery processes of the the case of Delta, indicates that there might be the potential organization. The results of the program were implemented in the case programs to improve the effectiveness of integra- in January 2002 and, as an outcome, the program produced tion and thus possibly improve the performance of the pro- a novel information system that supports the organization’s gram organization. The discussion of whether the renewed internal business processes. improvements in integration effectiveness are necessary— Case Beta is an ongoing development program in a large and whether they would enhance the performance of the international private sector-organization. The program was program radically or would only lead to minor improve- set up at the beginning of 2004 with an objective to improve ments and would thus require an excessive increase in and develop organizational capabilities related to the man- resource utilization—is left out of this study. Instead, it is agement processes in the customer interface. The planned assumed that the environment forces program organiza- duration of the program was initially two years. But after suc- tions to adopt forms of integration that are appropriate for cessful outcomes, the role of the program has shifted from a their purposes. short-term, temporally-limited development activity to a more stable form of organization of the development of capa- Analysis of Structural Complexity bilities by projects. In order to evaluate the structural complexity in the case Case Gamma represents an ongoing renewal program programs, two different indicators were analyzed: the num- in a large, international private-sector organization. The pro- ber of projects in the program and linkages among the proj- gram was initiated in spring 2004 with an objective to devel- ects. It was assumed that an increase in both number of op and launch a new operations management system in one projects and linkages among them will increase the struc- business unit of the organization. The system—planned to tural complexity of the program.

52 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Structural Alpha Beta Gamma Delta Characteristics

Maximum number of 6 5 3 40 concurrent projects Average number of 3,67 2,8 2 < 1 linkages from/to a project Number of linkages 2 to 4 1 to 4 2 0 to 2 from/to a project

Description of the Network of linkages Network of linkages Network of linkages Network of linkages inter-project linkage between projects between projects between projects between projects network of the program dense, interdependen- relatively dense, very dense, sparse, cies between projects interdependencies interdependencies interdependencies mostly reciprocal, high between projects between projects between projects level of required mostly sequential, sequential, moderate mostly pooled, low integration moderate level of level of required level of required required integration integration integration

Table 1: Characteristics of structural complexity in the case programs

First, because the number of projects may vary during the Case Gamma is the smallest program of the four cases execution of the program, it was decided that the number of in terms of number of projects. The structure of the link- projects refers to the maximum number of concurrent projects age network is very dense because the program represents in the program. Second, in order to map the network of linkages a coherent entity. The execution of each project is depend- among different projects, program managers and project man- ent on the others. Dependencies among the projects are agers in each case were asked to draw the organization of the sequential in nature. Each project in the program provides program with different projects. In addition, they were asked to some input for other projects, but at the same time oper- indicate with arrows existing linkages among the projects. The ates as an independent entity. concept of linkage was explained to refer to interdependency The cases Alpha and Beta are approximately the same between two projects that affects the execution of either or both size in terms of the number of projects. The network of of them. Furthermore, program managers and project managers linkages in both cases is relatively dense, case Alpha hav- were asked to explain in their own words the nature of each linking, however, on average more linkages per project than age. The overall picture of linkages among different projects in case Beta. The structure of linkage network in neither of each case program was formed by integrating different drawings the cases is as homogenous as in case Gamma. In case and explanations provided by the program manager and project Beta, dependencies among the projects are either pooled managers. Table 1 includes descriptive numbers related to both or sequential. In the case Alpha the dependencies between indicators of structural complexity. the projects are either sequential or reciprocal in nature. The analysis of structural complexity revealed that the case programs differ in the number of projects and in the Analysis of Uncertainty nature of the linkage network among the projects. Case The perceptional uncertainty related to execution of the Delta represents the largest program of the four cases in programs was measured through two concepts: novelty terms of number of projects. The network structure in this and analyzability. First, novelty was used to refer to the case is highly fragmented. The program is divided into three degree to which the program was different from the previ- different categories, reflecting the existing structure of the ous programs executed in the organization from the tech- parent organization. Furthermore, projects inside each cate- nological perspective, the structural perspective, and the gory either serve as isolated entities or form small clusters of resource or competence requirements perspective. two to five projects. The linkages among projects that exist Second, analyzability was used to refer to the degree of in different clusters are rare. Projects inside the same clusters understanding at the beginning of the program related are linked to each other through a common goal. However, to the program’s goals, schedule, budget, resource needs, the execution of different projects is relatively independent working methods, and internal dependencies. In this from each other. Thus, based on the categorization pro- study, it was assumed that the level of novelty is directly posed by Thompson (1967), it is concluded that the proportional to the perceived uncertainty, and that the dependency among the projects is mainly the pooled type in level of analyzability is reversibly proportional to the this program. perceived uncertainty.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 53 In each case, the level of novelty and analyzability was analysis proceeded iteratively, including various compar- measured through 10 7-point Likert scale statements. The isons among existing theories and empirical data. perceived level of novelty and analyzability related to each The integration mechanisms, obtained as a result of the case program is summarized in Figure 1. For more in-depth analysis, were further categorized into five different classes, information of averaged values for each statement and based on the formality of the mechanism and whether the respective range in answers, see Table 5 in Appendix 1. integration mechanism is personalized or impersonalized in The perceived levels of novelty and analyzability reveal nature. Categorization of the integration mechanisms is major differences among the case programs. In Case Alpha, modified from that proposed in the previous studies of Van the perceptual uncertainty is the highest. Case Alpha repre- de Ven et al. (1976), Kraut and Streeter (1995), and Tsai sents a highly novel program with a moderate level of ana- (2002). The respective categories for the observed integra- lyzability. Case Beta is highly analyzable in all measured tion mechanisms in this study are: formal group mecha- dimensions, and not very novel from technological, nisms, informal group mechanisms, formal personal resource or structural perspectives. Programs Delta and mechanisms, informal personal mechanisms, and formal Gamma are characterized by a moderate level of novelty and impersonal mechanisms (Table 2). analyzability. Group modes of integration refer to mechanisms in The wide range of answers (Table 5, Appendix 1) in which mutual adjustments occur in a group of occupants each case can be explained by several observed facts. First, (more than two) through meetings. Formal group mecha- each of the cases’ programs represents a complex collection nisms refer to meetings that are planned or scheduled, of different projects, some of which are more uncertain in whereas informal group mechanisms refer to autonomous, nature than others. Second, all case programs involved vari- unscheduled meetings. The analysis of data revealed three ous individuals representing different disciplines and differ- different types of formal group mechanisms: regular pro- ent organizational levels, such as shop-floor-level specialists, gram core team meetings, regular decision-making commit- professional project managers, middle-level managers, and tee meetings, and the collocation of program employees. even representatives of top management. Due to their dif- The third mechanism does not represent a meeting in its ferent positions in permanent organization, all these people deepest sense; however, it is included in this category have different perceptions of novelty and analyzability. because it was mentioned in several cases as an important Furthermore, different individuals have different experience mechanism that enables face-to-face communication on projects and programs that finally affect their concep- among several individuals. Three different types of informal tions of novelty and analyzability. group mechanisms for integration were found: autonomous unscheduled face-to-face meetings among several project Integration managers, facilitated informal face-to-face meetings among Integration Mechanisms and Their Importance several project managers, and informal network meetings. The in-depth analysis of interview data revealed 15 different The informal network meetings refer to situations that integration mechanisms. The analysis based on searching enable participants of the program organization to meet identifiable patterns (integration mechanisms) from the each other outside the program. Project management semi- transcribed interview data and from the filed notes. The nars and special focus group meetings are examples of net- observed patterns were coded using descriptive coding logic work meetings that were mentioned in the interviews. (Miles & Huberman, 1994). The codes changed and devel- Personal mode of integration refers to the mecha- oped during the analysis process, until the additional analy- nisms in which individual role occupants make mutual sis no longer refined the patterns. The process of data task adjustments through vertical or horizontal communication. Formal personal integration mechanisms are Low Moderate High those in which the use of individual role occupants as 7,00 integrators is planned, whereas informal personal integration mechanisms refer to integration through the Alpha High autonomous behavior of an individual role occupant. Three different formal personal integration mechanisms 5,00 were found. First, a program manager was used as a mes- Delta Moderate senger who participates actively in execution and deci- Gamma sion-making in different projects, serving as an integrator

Novelty 3,00 Beta who delivers information between projects. Second, the same employees and project managers were used as Low liaisons, participating in several projects in order to 1,00 ensure information sharing among projects. Third, an 1,00 3,00 5,00 7,00 external consultant was used as a coordinator, monitoring Analyzability the execution of projects and delivering information about critical points and boundaries among projects. Two Figure 1: Uncertainty in the case programs informal personal integration mechanisms were found:

54 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Formal Group Mechanisms 1 Regular program core team meetings 2 Collocation of core persons 3 Decision-making committees

Informal Group Mechanisms 4 Autonomous unregular face-to-face meetings (between several project managers) 5 Facilitated informal meetings between several project managers 6 Integration through informal interpersonal network meetings

Formal Personal Mechanisms 7 Integration via liaisons (project employees or project managers program coordinator) 8 Integration through messenger (program manager) 9 External consultant as a coordinator

Informal Personal Mechanisms 10 Direct contact between persons via e-mail or phone 11 Direct personal face-to-face contacts between employees or project managers

Formal Impersonal Mechanisms 12 Information exchange through reporting and formal documents 13 Integration through plans and schedules 14 Organizing: definition of roles, responsibilities 15 Information database

Table 2: Observed integration mechanisms and categories for integration mechanisms direct contact between project managers or project case) were summed. Thus, the importance of integration employees via e-mail or phone, and direct face-to-face mechanism i in case X (IMP) was calculated using Equation 1: contacts between project managers or project employees.

Formal impersonal integration mechanisms refer to X m X the use of a codified blueprint of action that is imperson- IMP(M ) = M (1) i ∑ j=1 ij ally specified. In-depth analysis of each case revealed sev- X eral integration mechanisms that fell into this category. where M ij denotes the importance of integration mechanism i The observed mechanisms are the use of formal docu- perceived by informant j in case X, and m denotes the num- ments and reports, formal plans and schedules, definition ber of informants in case X. Moreover, in order to compare of roles and responsibilities, and the use of standardized the results among the cases, the relative importance of the information systems such as a common database. integration mechanisms was calculated. The relative impor- In order to evaluate the importance of different inte- tance of integration mechanism i in case X (RIMP) was ana- gration mechanisms, the informants were asked to specify lyzed using Equation 2: the three mechanisms that they considered to be the most important from an information acquisition perspective, IMP(M X ) RIMP X = i and to rank these three mechanisms from most important i Max(IMP(M X ),..., IMP(M X )) (2) to least important. Furthermore, these perceptional evalu- 1 N ations of each informant were quantified by assigning the most important mechanism four points, the second-most- where N indicates the total number of integration mecha- important three points, and the third-most-important two nisms found in the study. points. All other mechanisms mentioned in the interview Table 6 in Appendix 2 includes the summary of the received one point each. results of the quantification of perceived importance of dif- In order to calculate the importance of each mechanism ferent integration mechanisms in each case. First, the results in each case, the points assigned each informant (in the of the analysis reveal that in each case several different

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 55 (8-11) integration mechanisms were used. In addition, the revealed that the interdependence among projects seems to relative perceived importance of different integration mech- increase the importance of autonomous unscheduled group anisms changed case by case. In all the cases, formal group meetings among project managers and the use of an exter- mechanisms were seen as the most important. Also, infor- nal coordinator. Moreover, interdependence among the mal group mechanisms and informal personal mechanisms projects decreased the importance of interpersonal net- were seen as highly important. Formal personal mecha- works, reporting and formal documents, and the use of proj- nisms and impersonal mechanisms were only seen as mod- ect employees or project managers as liaisons delivering erately important. Second, analysis reveals how the information between the projects. perceived uncertainty and structural complexity affects the perceived importance of different integration mechanisms. Discussion and Conclusion The summary of integration mechanisms, uncertainty, and This study provides novel insights on integration in a multi- structural complexity in each case is included in Table 3. project context. First, the in-depth analysis of four multiproject programs results in fifteen different mechanisms that The Effects of Uncertainty are used to ensure proper integration among different proj- The cross-case analysis between case Alpha (high novelty) and ects. Moreover, the different mechanisms seem to vary both case Beta (low novelty) reveals that the novelty of the program in their formality and to the extent to which they are per- seems to increase the importance of autonomous unsched- sonalized. Furthermore, the perceived importance of each uled meetings among projects managers, the use of project mechanism seems to be different in each case. managers and employees as a messenger delivering informa- Second, by expanding the emerging contingency think- tion among the projects, and the use of an external coordina- ing from a single-project context to a multiproject environ- tor enabling integration among the projects. Moreover, it was ment, the study demonstrates how the classical observed that, in case Delta, decision-making committees of contingencies, uncertainty and complexity, affect the per- the permanent organization were seen as more important ceived importance of observed integration mechanisms. The integration mechanisms than in other case programs. In addi- results of the study suggest that high uncertainty increases tion, the analysis revealed that the more novel the program is the importance of an informal group mode of integration to the organization, the less important are the formal plans and a formal personal mode of integration. In addition, the and schedules from the inter-project integration perspective. results reveal that the high level of uncertainty decreases the A somewhat surprising result is that the less novel the pro- importance of formal impersonal integration mechanisms gram is for the organization, the more important the direct and informal personal integration mechanisms. The results face-to-face contacts among program employees or project of the study partly support Daft and Lengel’s (1986) “theo- managers are seen as an integration mechanism. ry” of information requirements effects to the structural The effects of analyzability of the program for the design of the organization. However, the results also reveal importance of integration mechanisms did not appear very that, rather than emphasizing either integration mecha- clearly in this study because the cases represented only mod- nisms with high capacity to process information or integra- erate (Alpha, Gamma, Delta) and high (Beta) levels of ana- tion mechanisms with low capacity to process information, lyzability. However, the importance of two integration the use of various different integration mechanisms with mechanisms differ remarkably among the case Beta, in different information processing capabilities is important in which the analyzability of the program was high, and cases a program context. Alpha, Gamma, and Delta, which all represented moderate The analysis of complexity in the case programs also pro- analyzability. In case Beta, formal plans and schedules, and vided useful explanations for integration. Surprisingly, the direct personal contact among project managers or program distinctions among formal and informal, or personal, group, employees via e-mail or phone, seemed to be more impor- and impersonal did not seem to produce any significant dif- tant than in other less analyzable programs. ferences among the cases. Nevertheless, the analysis of individual mechanisms among the cases reveals that a large The Effects of Complexity amount of projects increases the importance of formal deci- The number of projects indicates the size of the program. In sion-making committees, and reporting and formal docu- case Delta, which represents a large program in terms of ments in integration. In addition, informal interpersonal number of projects, the interpersonal networks external to networks external to the program seem to be important the program were perceived as more important integration mechanisms to complement otherwise hierarchically-orient- mechanisms than in smaller programs. In addition, reported integration when the amount of projects is large. The ing and formal documents played a more important role results of the study support those of Mintzberg (1979), who than they did in other case programs. Furthermore, a high has acknowledged that a rich network of informal communi- number of projects seemed to decrease the importance of cation is, from an organization’s perspective, a highly impor- direct contacts among project managers or program employ- tant complement to otherwise relatively formal structures. ees as an integrating mechanism among projects. Finally, the study exposed some effects of inter-project Finally, interdependence among projects was highest in interdependence on the importance of integration mecha- case Alpha and lowest in case Delta. The cross-case analysis nisms. The in-depth analysis of the inter-project network

56 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Case Uncertainty Complexity Integration Strategy

Novelty Analyzability No. of Projects Interdependence Among Projects Alpha Very High Moderate Low High Integration strategy utilizes various channels of information delivery in order to assure proper inter-project integration. Formal and informal group mechanisms of integration are appreci- ated as highly important. The most important mechanisms for integration are regular program core team meetings and informal irregular ad hoc face-to-face meetings among several project managers. Formal and informal personal mechanisms of integration are perceived as moderately important. Program manager plays an especially important role as a liaison delivering information among different projects. Impersonal mechanisms of integration are low in importance.

Beta Low High Low Moderate Integration among different projects is primarily based on formal group mechanisms and informal personal mechanisms of integration. Formal program core team meetings and direct face-to-face or e-mail contact among project managers or project employees are highly important mechanisms in coordination. Impersonal integration mechanisms, especially integration through formal plans and schedules, is perceived as relatively important. Informal group mechanisms and formal personal mechanisms of integration are not perceived as important from the inter-project integration perspective.

Gamma Moderate Moderate Low Moderate Integration strategy is heavily based on group modes of integration. Regular program core team meeting and facilitated informal meetings among project managers are the most important integration mechanisms in this case. Informal personal integration through direct contact among projects via e-mail or phone was perceived as relatively important. Also integration through impersonal vehicles such as a common database was seen as moderately important. Use of liaisons, messengers or coordinators was not perceived as important from the integration perspective.

Delta Moderate Moderate Very High Low Integration is based on information delivery in parent organization’s permanent decision- making committees. This strongly hierarchy- based integration strategy is complemented with formal reporting practices that are perceived as moderately important. Informal interpersonal network meetings, and direct face-to-face contacts among projects support this mechanistic-oriented integration strategy. These mechanisms are also perceived as relatively important from the integration perspective.

Table 3: Inter-project integration mechanisms, uncertainty and complexity

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 57 within each case and among cases revealed that the high Science Quarterly, 27(3), 420–434. amount of interdependency seems to create a need for a Artto, K. A., & Dietrich, P. H. (2004). Strategic business two-fold structure. First, interdependence leads to the adop- management through multiple projects. In P. W. G. Morris & tion of a separate coordinator serving as a formal linkage J. K. Pinto (Eds.), The Wiley guide to managing projects. among different projects. In addition, interdependence Hoboken, NJ: John Wiley & Sons. among projects induces autonomous informal meetings in Clark, K., & Fujimoto, T. (1991). Product development per- order to respond to local emerging problems. formance: Strategy, organization, and management in the world The analysis of structural contingencies in multipro- auto industry. Boston: Harvard Business School Press. ject environments reveals that not all organizational theo- Cleland, D. I. (2002). Project management—Strategic ries apply in a complex multiproject context. Neither are design and implementation. New York: McGraw-Hill. all ideas from the traditional single-project context appli- Cohen, M., & Regan, R. (1996). Management Internal cable to programs. The study provides valuable insights for Consistency in Technology Intensive Projects. both researchers and program managers challenged to Competitiveness Review, 6(1), 42–58. introduce strategic or large-scale changes in organizations. Daft, R. L., & Lengel, R. H. (1986). Organizational First, the study reveals that, in addition to the formal plan- information requirements, media richness and structural ning perspective, emphasized in managerial guidebooks, design. Management Science, 32(5), 554–571. informal mechanisms are highly essential from an integra- Donaldson, L. (2001). The contingency theory of organization perspective. Moreover, the study demonstrates that tions. Thousand Oaks, CA: Sage Publications. proper mechanisms for inter-project integration are Eisenhardt, K. M. (1989). Building theories from case dependent on uncertainty and the complexity of the pro- study research. Academy of Management Review, 14(4), gram. Thus, “one-size-fits-all” solutions do not lead to 532–550. desired results. It should be noted that, in all types of pro- Gareis, R. (2004). Management of the project-oriented grams, the proper integration strategy should allow the use company. In P. W. G. Morris & J. K. Pinto (Eds.), The Wiley of various alternative mechanisms for information deliv- Guide to Managing Projects. Hoboken, NJ: John Wiley & Sons. ery. The key question is: what kinds of mechanisms are the Gittell, J. H. (2001). Supervisory span, relational coor- most essential in each situation? dination and flight departure performance: A reassessment When interpreting the results of this study, it should be of post-bureaucracy theory. Organization Science, 12(4), noted that this research focused on intra-organizational 467–482. development programs. Thus, other types of programs—e.g., Gittell, J. H. (2002). Coordinating mechanisms in care research programs or policy programs—might require dif- provider groups: Relational coordination as a mediator and ferent types of integration mechanisms that have not been input uncertainty as a moderator of performance effects. observed in this study. Furthermore, it is essential to under- Management Science, 48(11), 1408–1426. stand that this research is explorative in nature and that the Hage, J., Aiken, M., & Marrett, C. B. (1971). logic of analysis is based on an in-depth understanding of a Organization structure and communications. American relatively small number of cases. Thus, the results of the Sociological Review, 36(5), 860–871. study should be seen as propositions rather than statistical- Kraut, R. E., & Streeter, L. A. (1995). Coordination in soft- ly verified laws. Finally, more research is clearly needed to ware development. Communication of the ACM, 38(3), 69–81. test the results of this study and to explore additional con- Lawrence, P. R., & Lorsch, J. W. (1967). Organization tingencies and respective management strategies in a multi- and environment. Managing differentiation and integration. project context. An important area of further research is the Boston: Harvard Business School Press. exploration of the way that the importance of different inte- Miles, M. B., & Huberman, A. M. (1994). An expanded gration mechanisms change and develop during the pro- sourcebook: Qualitative data analysis. Thousand Oaks, CA: grams’ “lifecycle.” Sage Publications. Milosevic, D. Z., & Patanakul, P. (2002). The impact of References standardized project management: New product develop- Adler, P. S. (1995). Interdepartmental interdependence ment versus software development projects. Proceeding of PMI and coordination: The case of the design/manufacturing Research Conference 2002. 14-17 July 2002, Seattle, WA, USA. interface. Organization Science, 6(2), 147–167. Mintzberg, H. (1979). The structuring of organizations. Albernathy, F. H., Dunlop, J. T., Hammond, J. H., & Englewood Cliffs, NJ: Prentice-Hall, Inc. Weil, D. (1999). A stitch in time: Lean retailing and the trans- Nidumolu, S. R. (1996). A comparison of structural formation of manufacturing—Lessons from the apparel and tex- contingency and risk-based perspectives on coordination in tile industries. New York: Oxford University Press. software-development projects. Journal of Management Andres, H. P., & Zmud, R. W. (2001). A contingency Information Systems, 13(2), 77–113. approach to software project coordination. Journal of Pellegrinelli, S. (2002). Shaping context. The role and Management Information Systems, 18(3), 41. challenge for programmes. International Journal of Project Argote, L. (1982). Input uncertainty and organizational Management, 20(3), 229–233. coordination in hospital emergency units. Administrative Pellegrinelli, S. (1997). Programme management:

58 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Organizing project-based change. International Journal of product innovation. IEEE Transactions on Engineering Project Management, 15(3), 141–149. Management, 47(1), 74–87. Perlow, L. A., Gittell, J. H., & Katz, N. (2004). Contextualizing Thompson, J. D. (2003). Organizations in action: Social patterns of work group interaction: Toward a nested theory of science bases of administrative theory. New Brunswick, NJ: structuration. Organization Science, 15(5), 520–236. Transaction Publishers. Ribbers, P. M. A., & Schoo, K. C. (2002). Program man- Tsai, W. (2002). Social structure of “coopetion” within a agement and complexity of ERP implementations. multiunit organization: Coordination, competition, and Engineering Management Journal, 14(2), 45–52. intraorganizational knowledge sharing. Organizational Shenhar, A. J. (2001). One size does not fit all projects: Science, 13(2), 179–190. Exploring classical contingency domains. Management Van de Ven, A. H., Delbecq, A. L., & Koenig, R. (1976). Science, 47(3), 394–414. Determinants of coordination modes within organizations. Shenhar, A. J., & Dvir, D. (1996). Toward a typological American Sociological Review, 41(2), 322–338. theory of project management. Research Policy, 25(4), Vereecke, A., Pandelare, E., Deschoolmeester, D., & 607–632. Stewens, M. (2003). A classification of development pro- Tatikonda, M. V., & Rosenthal, S. R. (2000). Technology grammes and its consequences for programme manage- novelty, project complexity, and product development proj- ment. International Journal of Operations & Production ect execution success: A deeper look at task uncertainty in Management, 12(10), 1279–1290.

PERTTU DIETRICH works as a researcher and project manager in BIT Research Centre in Helsinki University of Technology. Mr. Dietrich has several years experience on research and development work with public and private sector organizations. His research interests include the following themes: project management, multiproject management, program management, coordination theory, contingency theory, innovation management, and strategy management.

Appendix 1

Table 4: Perceived effectiveness of inter-project integration in the case programs

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 59 Appendix 1 (cont.)

Uncertainty Alpha Beta Gamma Delta

No. Statement Average Range Average Range Average Range Average Range NOVELTY 1 The outcome of the program differed 6,33 4 to 7 3,00 1 to 6 3,8 2 to 5 4,00 2 to 6 remarkably from the previous programs in the organization from the technological point of view 2 The execution of the program differed 5,83 3 to 7 2,50 1 to 6 2,6 2 to 3 3,56 1 to 5 remarkably from the previous programs in the organization from the technological point of view 3 The resource and completence needs of 6,00 5 to 7 2,63 2 to 6 3,4 2 to 5 4,22 1 to 6 the program differed remarkably from the previous programs in the organization 4 In the program there were much more 6,17 5 to 7 3,13 1 to 7 5,2 4 to 6 4,22 1 to 6 dependencies between projects compared to previous programs in organization Average 6,08 2,81 3,75 4,00 ANALYZABILITY 5 In the beginning of the program there was 5,00 4 to 6 4,75 2 to 7 5,00 2 to 6 3,56 2 to 6 understanding about the dependencies (related to projects’ execution) between different projects in a program 6 In the beginning of the program there was 4,67 2 to 6 5,13 3 to 6 5,00 2 to 6 3,67 2 to 7 understanding about the work methods to be used in projects 7 In the beginning of the program there was 6,33 5 to 7 6,00 4 to 7 5,60 3 to 7 5,33 2 to 7 defined a clear schedule for the projects in the program 8 In the beginning of the program there was 2,60 1 to 4 6,38 5 to 7 5,20 2 to 7 3,78 2 to 7 defined clear budget for the projects in the program 9 In the beginning of the program there 5,50 4 to 7 4,57 3 to 6 4,40 2 to 7 4,56 2 to 7 were defined measurable goals defined for the projects in the program 10 In the beginning of the program the 4,00 2 to 5 4,86 2 to 7 3,20 2 to 6 3,33 2 to 7 resource needs were defined for the projects in the program Average 4,68 5,28 4,73 4,04

Table 5: Perceived level of uncertainty in the case programs

60 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Appendix 2

No. Observed Integration Mechanisms Alpha Beta Gamma Delta

Formal Group Mechanisms 1 Regular program core team meetings 1,00 1,00 1,00 - 2 Collocation of core persons 0,13 0,44 0,08 - 3 Decision-making committees - - - 1,00 Informal Group Mechanisms

4 Autonomous unregular face-to-face meetings (between several project managers) 0,81 - 0,31 - 5 Facilitated informal meeting between several project managers - 0,06 0,92 - 6 Integration through informal interpersonal network meetings - - 0,08 0,48 Formal Personal Mechanisms

7 Integration via liaisons (project employees or project managers program 0,06 0,19 0,15 0,28 coordinator) 8 Integration through messenger (program manager) 0,50 - 0.23 0,20 9 External consultant as a coordinator 0,31 - - - Informal Personal Mechanisms

10 Direct contact between persons via e-mail or phone 0,56 0,69 0,54 0,16 11 Direct personal face-to-face contacts between employees or project managers 0,13 0,81 0,31 0,40 Formal Impersonal Mechanisms

12 Information exchange through reporting and formal documents 0,31 0,31 - 0,60 13 Integration through plans and schedules - 0,50 0,08 - 14 Organizing: definition of roles, responsibilities 0,13 0,06 - - 15 Information database - 0,13 0,46 0,36

(*) Number 1,00 indicates the most important integration mechanisms in each case and numbers < 1,00 indicate the relative importance of each other mechanism in the case, “-” indicates that integration mechanism was not observed in the case.

Table 6: Relative perceived importance of integration mechanisms in the case programs (*)

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 61 LONGITUDINAL ANALYSIS OF PROJECT MANAGEMENT MATURITY

MARK MULLALY, School of Business, University of Alberta

Introduction ABSTRACT he assessment of organizational capabilities is a core dimension of organizational learning and improvement. As organizations strive to attain and This paper examines and identifies core T retain competitive advantage, an understanding of their capabilities and how dimensions of assessment frameworks, including five core requirements for con- these compare with competitors and best-in-class organizations is essential. ducting assessments, two key processes Within the project management sphere, assessment frameworks have become of assessing organizations (audit and increasingly prevalent, and, in particular, the development and application of self-assessment), and two dimensions of project management maturity models. The majority of frameworks have been improving performance (delivering data developed in the last three to five years. and applying data). It discusses the evolution of using maturity models to assess Although there has been a growing emphasis and focus on assessing project organizational capabilities and the devel- management capabilities, there is little data that is available to provide an objec- opment of maturity models to assess tive understanding of the current state of practice. What data does exist tends to be project management competencies. It proprietary in nature, and as a result there is little information in the public then outlines a five-level project manage- domain that organizations can utilize to understand and benchmark their capa- ment maturity model that the authors used to assess the way 550 international bilities. More particularly, there is very little longitudinal data of project manage- organizations practice project management capabilities and performance over time. Because principles of organizational ment. The paper lists the challenges, development require the ability to progressively evaluate progress (Nielsen & advantages, and disadvantages of using Kimberly, 1976), lack of a longitudinal view of project management capability this model; it identifies the practices syn- seriously curtails the ability of organizations to identify key drivers of project man- onymous with improvements in demonstrated maturity. It also compares the agement improvement. results of data collected since this bench- This paper provides a review of the literature associated with organizational marking study’s inception, results that assessment, identifying the core dimensions that assessment frameworks need to show underlying project management address. It discusses the rise of maturity models as a means of assessing specific trends, such as changes in organizational functional capabilities within organizations, and explores the history and devel- capabilities and performance. It reviews the impact of these trends on the studied opment of one particular model for assessing project management maturity. organizations and the way they manage Although the model discussed in this paper is proprietary in nature, the paper their projects. It concludes by detailing explores the results of a series of benchmarking studies that have placed assess- four key—and unexpected—results. ment results in the public domain. Most importantly, it provides an initial longitudinal analysis of the changes in capabilities and performance of organizations Keywords: project management maturity; organizational capabilities; organiza- in managing their projects. tional assessments Approaches to Organizational Assessment ©2006 by the Project Management Institute Organizations continually engage in assessment activities. According to Nielsen Vol. 36, No. 3, 62-73, ISSN 8756-9728/03 and Kimberly (1976), this is a product of the organizational need to rationally search for opportunities for continued improvement, assign priorities, and make

62 AUGUST 2006 PROJECT MANAGEMENT JOURNAL decisions. To support this assessment, they identify five core and resulting actions resulting from delivery of the survey requirements for assessment: data (Conlon & Short, 1984). Conlon and Short stated that • The availability and interpretation of information, in the way in which information is delivered is an important a form that is of use and at the time that it is required determinant of the effectiveness of an assessment. • An understanding of what has been assessed, with a Effectiveness has been found to increase based on member clear understanding of the goals of the assessment involvement, where the receiving audience is prepared for and a defined knowledge of the outcomes and conse- the assessment feedback and where they are able to under- quences that result from the assessed resources and stand and take action on the data received. procedures A large number of assessment frameworks have been • The availability of relevant and appropriate measures adopted organizationally in recent years. Audit frameworks of the consequences being assessed are generally tied to external quality standards such as the • A data collection strategy by which to gather the various versions of ISO 9000. Tools for self-assessment are appropriate measures also rooted in total quality management, generally based on • An assumption of the cause-and-effect relationships quality award criteria, such as the Malcolm Baldrige that define the beliefs and support the decisions to be National Quality Award (MBNQA) and the European made as a result of the assessment. Quality Award (EQA). Although these provide general frameworks for evaluating organizational effectiveness, the Because these cause-and-effect relationships are often risk with any assessment is that it will lead to a long list of oversimplified, and the causal linkages are generally more strengths and weaknesses that are not tied to any specific complex than the decisions and beliefs imply, from the per- strategies leading to sustainable competitive advantage spective of Nielsen and Kimberly they are nonetheless nec- (Duncan, Ginter, & Swayne, 1998). Furthermore, customers essary in order to make meaningful interpretations. of assessment frameworks need to cast a critical eye on what The increasing prevalence of knowledge work creates each assessment offers, recognizing that their underlying additional challenges in assessment (Tuttle & Romanowski, approaches and ability to support differentiation vary sig- 1985). Although the underlying performance of an organi- nificantly (Biazzo & Bernardi, 2003). zation can be measured by five dimensions—efficiency, productivity, effectiveness, quality, and quality of work—these Role of Maturity Models are less easily measured as the complexity of work increases Models such as those used by the MBNQA and EQA support and its tangible nature declines. Direct outcomes, where an overall assessment of organizational excellence. Similar there is a direct relationship between outcome and output, forms of self-assessment addressing specific functional areas are more reliant upon measures of efficiency, productivity, of concern have been popularized through the various and quality. Indirect outcomes, where there is a greater vari- maturity models. The concept of maturity models has been ability of potential outputs, and greater complexity in familiar for some time; however, their popularization as a choosing the right output for the desired outcome, places a means of assessment has been more recent. One of the best- much greater importance on the assessment of effectiveness, known maturity models, originally referred to as the capa- and to a lesser extent productivity and efficiency. bility maturity model for software (CMM-SW) and Although the core emphasis of all assessments is on developed by the Software Engineering Institute (SEI) of organizational learning (Hellsten & Wiklund, 1999), the Carnegie Mellon University, has significantly increased majority of assessment frameworks draw on the underlying awareness and acceptance of the concept. Originally principles of total quality management, rooted in the plan- released in 1991, the CMM-SW popularized the concept of do-check-act cycle of Deming (1993). These assessments are maturity models as consisting of a series of levels across a typically divided into two key assessment processes: audit number of capability areas (Humphreys, 1992). Since the and self-assessment (Karapetrovic & Willborn, 2001). Audits popularization of the CMM and its siblings by SEI, a variety collect and compare data against a reference standard, eval- of maturity models have been developed to support a range uating the degree to which the criteria have been fulfilled, of functions, including innovation (Aiman-Smith, whereas self-assessments are designed to evaluate the Goodrich, Roberts, & Scinta, 2005), strategic management strengths, weaknesses, and opportunities for improvement (De Vries & Margaret, 2003), contract management (Garrett against a number of dimensions. Audits are primarily & Rendon, 2005), and even more specific purposes such as designed to support an external driver of compliance, the use of enterprise resource planning software (Holland & whereas self-assessments are more typically internally Light, 2001). focused on improvement. The application of maturity models to project manage- Karapetrovic and Willborn suggested that not only can ment is comparatively recent. Despite their relative novelty, assessments provide a means of performance measurement, a large number of models have been released in recent years but by their nature they are also enablers of improved per- (Cooke-Davies & Arzymanow, 2003; Hillson, 2003; Ibbs & formance, particularly with respect to self-assessments. To Kwak, 2000; Jachimowicz, 2003; Sawaya & Trapanese, 2004; be effective in improvement, however, requires two dimen- Skulmoski, 2001). Many of those developed have adopted sions: delivery of the survey data itself, as well as the impact the framework and structure originally established by the

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 63 CMM, with five levels and a number of capability areas as results over the six years for which data exists provides the focus for assessment. These maturity models have vary- insights into underlying trends and the impacts of these ing levels of formality, and there is little documentation in trends on organizations. the public domain regarding their structure, contents, The results of this benchmarking effort have been pub- assessment approach, or results. Even less information is lished in the public domain since the first year, with the available as to the degree to which maturity models actual- executive summary of each year’s research published on ly support improvement in project or organizational results. Interthink’s web site (www.interthink.ca). More results The most widely known study of the relationship between included in this analysis (Mullaly, 2004) reflect the consoli- maturity and organizational results (Kwak & Ibbs, 2000) dated findings of the 2003 study. Until now, the results have demonstrated no statistically significant correlation between been limited to the findings for the year in which the bench- process maturity and project results, despite the lack of hard marking survey has been conducted, with no longitudinal results an anecdotal link was claimed. analysis of the resulting data. This paper provides this lon- In evaluating the use and effectiveness of project man- gitudinal perspective by evaluating the changes in results agement maturity models, Jugdev and Thomas (2002) and their underlying causes over the period that this bench- found that the claimed correlation between process capa- marking has been conducted. bility and project success of many maturity models has not been substantiated. For prospective customers seeking a rel- The Maturity Model evant assessment framework, the failure of any one model The findings within this paper are derived from a maturity to achieve widespread acceptance is equally problematic. model initially developed in 1993. Similar in structure to Building upon the observations of Duncan et al. (1998), many others, this model originally drew its inspiration from the larger concern is of the ability of project management the framework and assessment approach of the CMM maturity models to offer a demonstrable means of compet- (Humphreys, 1992). The structure of five levels that is itive advantage. Although it can be argued that maturity defined within the CMM has been adapted in order to pro- models have in fact helped to elevate the discussion of proj- vide relevance for project organizations, resulting in the fol- ect management and raised awareness of its contribution to lowing descriptors: organizational success (Jugdev & Thomas, 2002), there is • Level 5. A fully mature project organization, with still very little empirical information currently available to processes consistently applied throughout the organi- support their use. No recognizable standard has emerged to zation as part of the overall management process. assess project management practices, and in particular there • Level 4. A mature project management process applied is little to no evidence-based data to support assessment consistently on all projects, with project management and improvement using the available models What infor- recognized as a formal management discipline. mation does exist regarding organizational capabilities • Level 3. An organization with a defined and integrat- tends to be proprietary and therefore not publicly available, ed project management process that is consistently and in particular there has typically been no longitudinal applied on each project. data available. • Level 2. Some project management capabilities defined, but not consistently applied. One Approach to Assessing Project Management Maturity • Level 1. A fully ad-hoc project management capabili- Although project management maturity models have not ty; no consistent or repeatable processes. necessarily fully demonstrated their contribution, any insights that can be derived can still be of some value. This As well as the levels of maturity previously defined, the paper provides a comparison over time of how organizations other dimensions of assessment are comprised of 12 capa- have been assessed against one project management maturi- bility areas that reflect the aspects of project management ty model. The value of conducting this longitudinal analysis practice within organizations being assessed. Within each is that it provides an initial understanding of how both the capability area, a number of capabilities are defined that application of project management in organizations may represent how each might be carried out in an organiza- have changed over time, and the corresponding impact these tional context. Within each capability, a number of practices changes have had on the organizations making them. are identified that align with each of the levels within the The data in this paper derives from a benchmarking ini- maturity model. tiative conducted by a project management consulting com- One of the early challenges recognized in using the pany since 1997. Over the six years that the organization has model was the lack of comparative information to support conducted the benchmarking, the results have increasingly assessment of an organization’s practices against other suggested a link between the improvement of project man- organizations, as well as the difficulty in quantifying the agement capabilities in organizations and the delivery of impact of improvements on an organization’s practices. In successful project results. As well, the findings for each year an effort to provide a context by which comparisons could have identified practices that have had a strong correlation be better established, the survey instrument was developed with improvements in demonstrated maturity as defined to support assessments against the model. Subsequently, within the underlying maturity model. Comparing the public benchmarking activities were initiated by the firm.

64 AUGUST 2006 PROJECT MANAGEMENT JOURNAL The model offers a number of advantages and disad- ity awards, the model adopts a similar framework in vantages that need to be understood in evaluating the data that it is not simply prescriptive as would be defined presented here. by a level 1 or 2 instrument and it allows organiza- • The model itself is proprietary, and has never been tions to evaluate their capability against a range of published. Although the benchmarking results have potential practices. It is also not as open and fluid as been placed in the public domain, there has been no a level 4 instrument in that it does not allow for diag- empirical verification or validity assessment of the nosis or design without relying on judgment criteria, constructs within the model, which were developed and it certainly cannot be associated with openness of primarily through expert opinion, analysis of existing the level 5 practices—the majority of project manage- project management standards, and testing and vali- ment organizations would not have sufficiently struc- dation of the principles with consulting customers. tured processes to support the causal analysis this However, the results of more than 60 organizational level requires. assessments have been reviewed directly with project management stakeholders, and findings of the model Even taking into account these factors, the use of the have shown a high level of face validity in describing model as a framework to assess project management matu- the practices and capabilities. rity over time still offers significant value in that it has been • The maturity model aligns with the principles of a used to collect a large base of benchmark data, predomi- self-assessment model as described by Karapetrovic nantly from North American organizations, over a period of and Willborn (2001). It does not offer a prescriptive six years. Within this base of data, a number of organiza- model of project management, but allows organizations have participated over several years, offering further tions to evaluate their relative strengths and weak- insights into the impact of longitudinal changes within a nesses against a range of practices. specific organization as well as the overall trends of the • The focus of the model is one of promoting under- study as a whole. standing and improvement. As per the analysis of Tuttle and Romanowski (1985), project management Participants aligns with their definition of an indirect outcome; Data presented in this paper is derived from a public bench- therefore, the model has been designed primarily as a marking project. This effort was initially limited to organi- measure of effectiveness. Productivity and efficiency zations in Canada, but was subsequently expanded to North are not factors that are measured within the model, America in 1998 and worldwide in 2001. More than 550 and both quality and quality of work life are evaluat- organizations and 2,500 individuals have participated in the ed only to a lesser degree. study since its launch in 1998. • The assessment of effectiveness is not as optimally Participant organizations were solicited through direct linked to project success as would generally be desired. mail, e-mail, advertisements, and editorial articles to con- Although the underlying assumption of the model is a tribute to the study. No compensation was provided to correlation between process maturity and project suc- organizations for participating, although organizations with cess, the benchmarking results to date are primarily more than 10 participants were offered a complementary based upon self-report data. Only a subset of respons- customized briefing of their individual results. For the first es has been verified through interviews and follow-up two years of the study, the Project Management Institute reviews of defined and applied practices. Because of (PMI) made available its mailing list within Canada to the difficulty of normalizing project measures across invite participants; in subsequent years, direct mailings were organizations, project success as measured by delivery primarily drawn from the firm’s contact database. on time, budget, effort estimate, customer expectation, Organizations were encouraged to have a cross-section of and customer satisfaction have been evaluated by self- participants contribute to the survey, including project man- reported ranges of result (>-25%, -25% to -10%, -10% agers, sponsors, stakeholders, and team members. This diver- to +10%, +10% to +25% and >+25% of target). sity helped ensure a more balanced view of how projects are • Although the model embeds an understanding and conducted, capturing both the perceptions of project man- assumptions of cause and effect, these are inherent in agers in describing their activities and the degree to which the model itself and are applied equally to all organi- these activities are actually observed by other participants. zations. As per Nielsen and Kimberly (1976), these In general, participant organizations were those who are not adapted or tailored to the individual contexts, already had some form of organizational project manage- needs, or goals of individual organizations, although ment capability being developed, drawn from a wide array the nature of the results do enable organizations to of industries. define their specific project management improvement goals relative to the reported results. Procedures • The model could be considered analogous to a level 3 In responding to the survey, participants completed a 135- assessment model as described by Biazzo and question multiple-choice survey, describing practices they Bernardi (2003). Although not a framework for qual- utilized or observed in their most recent project.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 65 Participants could provide more than one answer, but their project management staff on the year 2000 remedia- were asked to rank multiple answers relative to the degree tion efforts. Since this time, the number of individual par- to which the described practice reflected how projects were ticipants has been growing, while the number of actually managed. organizations remains fairly static—this is largely a result of The survey results were correlated to the maturity a larger number of participants per organization respond- model, so that the described practices were each related to a ing, with organizations being encouraged to provide multi- particular level within each of the defined process capabili- ple respondents in order to provide as relevant a picture of ties. Not all capabilities have practices associated with all their processes and capabilities as possible. levels, and multiple practices described within a question For participants, overall experience in project man- could be associated with the same level. The results for each agement has remained relatively static year after year, level were averaged using a weighted formula, so that for an with results consistently reflecting a broad range of expe- organization to be ranked at a level 2, for example, all prac- rience in managing projects. On average, 32.5% of tices associated with level 1 and level 2 at a minimum need- respondents report less than 5 years experience, 27.2% ed to be met. report between 5 and 10 years, and 40.3% report more Where organizations had more than one respondent, than 10 years of experience. the results of each respondent were averaged to produce the What has seen a change from year to year has been the overall results for the organization. As well, the standard training experience reported in project management. deviation of responses was assessed to understand the rela- Although the numbers without formal training and with a tive consistency and range of differences in responses across bachelor’s degree in project management or a related disci- participants in each individual organization. The composite pline have remained fairly static, there has been a slight organizational results were then used for all subsequent increase in those reporting a master’s degree in project man- data analysis in presenting study findings. agement or a related discipline. There has also been a rela- In addition to describing their project management tively large increase in those reporting a certificate in project practices, participants also responded to a number of demo- management, rising from low of 11.4% in 1998 to a high of graphic questions associated with them as a person, their 26.3% in 2002, with an average of 23.5% of respondents for organization, and their most recent project. In all instances, the last three years. participants were asked to respond to the survey in the context of this project in order to best reflect the most current Organizations practices within their organization. The size of participating organizations has remained extremely consistent from year to year, with the majority of Longitudinal Analysis organizations (greater than 70% in all but one year) having Demographic Changes more than 1,000 employees. The year 2001 also saw a high- In understanding changes in results over time, it is first er percentage of smaller organizations. important to understand what changes have been reflected The structure of participating organizations has shown in the respondents to the study and the organizations they a much greater degree of variation than other dimensions. belong to. The following sections provide a summary of the While incorporated, publicly-traded organizations have rou- changes and trends in demographic information. tinely represented a significant percentage of participants, privately-held organizations have declined in overall pres- Respondents ence in the study from an initial proportion of nearly a third Table 1 shows that the number of respondents year over year of all organizations to just above 10% in the most recent two has been reasonably consistent, with the exception of the years. Government participation has shown the most dra- year 2000. An average of approximately 300 participants per matic shift. Although 1999 saw a significant proportion of year and 70 organizations per year contributed to the study governmental organizations participating, this rapidly (omitting the anomalous years from the average). The fig- decreased in 2000 but has since grown in size every year, ures for 2000 were much lower as a number of organiza- with the most recent year representing over half of all organ- tions declined to participate, citing in particular the focus of izations when included with crown corporations.

# Respondents # Organizations Changes to Overall Maturity Anecdotally, organizational emphasis on project manage- 1998 280 67 ment has been increasing. Intuitively, what would be expect- 1999 337 63 ed as a result is relatively stable to somewhat increasing 2000 152 22 levels of maturity being exhibited in organizations as they 2001 246 89 continue to invest in developing and improving their proj- 2002 342 96 ect management capabilities. However, analysis of the overall levels of maturity for 2003 579 89 organizations has shown a surprising and counterintuitive Table 1: Summary of respondents by year shift, as shown in Figure 1. For the first two years of analy-

66 AUGUST 2006 PROJECT MANAGEMENT JOURNAL 100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0% 1998 1999 2000 2001 2002 2003 Overall Level 5 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Level 4 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Level 3 6.0% 4.8% 4.5% 1.1% 2.1% 0.0% 2.6% Level 2 64.2% 65.1% 31.8% 29.2% 29.2% 28.1% 39.9% Level 1 29.9% 30.2% 63.6% 69.7% 68.8% 71.9% 57.5%

Figure 1: Overall levels of maturity of participating organizations sis, the proportions of organizations assessed overall at lev- What is most significant in these findings is the absence els 1, 2, and 3 were relatively constant—approximately 30% of any gradual shift in results—the 1998 and 1999 results of organizations were at a level 1, while two-thirds of organ- are remarkably consistent, as are those from 2000 through izations reflected a level 2 capability (all capability areas 2003, but there is a significant drop in maturity from the were assessed at or above level 2) and approximately 5% of 1999 to the 2000 results that remains through all subse- organizations were assessed at level 3. Since 2000, there has quent years. The challenge is to understand the reason for been a marked increase in organizations that were assessed this significant shift. Within the demographics of respon- at level 1, with the percentage of organizations being greater dents and respondent organizations, there is minimal indi- than 60% for all subsequent years. In the period between cation that any real change in maturity should be 2001 and 2003 the average percentage of level 1 organiza- reasonably expected. While there was a decline in the num- tions was 70.1%. ber of organizations participating in the study in 2000, sub- As a result of the increase in organizations at level 1, sequent years have seen participation levels similar to those there has been a corresponding decrease over time in seen in 1998 and 1999. As well, although there has been an organizations evaluated at level 2 or above. Level 3 organi- increase in government participation and a decrease in par- zations declined to 0% by 2003; the average proportion of ticipation by privately-held incorporated organizations, this level 3 organizations in the first three years was 5.1%, com- has been a progressive change year after year, which does pared with an average of 1.1% in the last three years. The not explain the rapid change or the subsequent static profile proportion of level 2 organizations for the period has been of maturity in each year. a surprisingly steady, with close to 30% of organizations at There have also been no significant changes to the level 2 since 2000. assessment survey or the means by which survey results are The trends viewed at a macro level in terms of overall evaluated that would explain this shift, as mentioned previ- assessment results are also reflected when viewing the over- ously. There have been few changes to the survey, and all of all results by year against each of the capability areas. The those changes that did occur in the period being explored results for 1998 and 1999 show a clear difference of approx- were clarifications of terminology in a small number of imately one half a level greater maturity overall over the individual questions rather than a significant restructuring results for all subsequent years. Beginning in 2000, the aver- of the survey. age results by capability area are surprisingly consistent, par- The source of the change appears to be external, and ticularly in the process capability areas—the variation attributable to one of two factors: either there has been a between years is typically less than one fourth of a level material change in how organizations approach project overall from 2000 to 2003. management since 2000, or there has been a significant shift

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 67 5.00

4.00

3.00

2.00

1.00 Program Project Project Plan Project Project Project Risk Technology Environment Organization Resource Contract Initiation Initiation Sched & Management Tracking Reporting Management Management Management Budget Average 2.60 2.67 2.55 2.53 2.13 2.37 2.40 2.28 2.23 2.65 2.26 2.25 1998 3.07 3.10 3.97 2.93 2.47 2.70 2.72 2.51 2.61 2.95 2.60 2.64 1999 3.22 3.17 3.03 3.04 2.48 2.67 2.77 2.55 2.46 3.04 2.56 2.56 2000 2.48 2.51 2.45 2.39 2.08 2.43 2.10 2.24 2.20 2.60 2.28 1.88 2001 2.35 2.47 2.37 2.35 1.97 2.27 2.26 2.30 2.15 2.55 2.16 2.11 2002 2.25 2.49 2.41 2.37 2.02 2.20 2.31 2.14 2.09 2.54 2.14 2.19 2003 2.31 2.44 2.28 2.35 1.97 2.18 2.35 2.08 2.03 2.33 2.02 2.01

Figure 2: Overall maturity by capability areas of participating organizations

in the composition of organizations participating in the has been a less effective and reliable means of reaching study since that time. The challenge in interpreting the prospective respondents and the primary means of soliciting underlying reason for this shift is that there is minimal data participants is through direct mail to an extensive database to support resolving the question one way or the other. of contacts maintained by the consulting firm. While a significant change in overall approach to proj- As discussed previously, the data does not support ect management could be plausibly the result of changes in being able to effectively understand the reason for this shift the business environment—as a reaction to the manage- in overall maturity. Further analysis of future years, follow- ment efforts associated with the year 2000 as well as a ing up with previous participants and longitudinal assess- response to the economic downturn that occurred through- ment of individual companies within the data set would all out 2001 for many regions—the data appear too neat in help to better understand what is influencing the changes terms of the suddenness of the change and its subsequent being observed within the results. consistency to support this as an explanation. A more likely and reasonable interpretation is a change Changes Within Organizations as to the organizations participating in the benchmarking While not all organizations that have contributed to the study. Although some organizations have been consistent benchmarking effort firm have participated in multiple years, supporters and participants of the benchmarking study from there are a sufficient number of participating organizations the outset, for the most part there tends to be a turnover in that have done so to enable an initial longitudinal analysis at organizations participating from year to year of approxi- this level. The following sections provide comparisons across mately 50%. For the first two years of the study, participants multiple years for three different organizations that have par- were solicited in part through PMI’s mailing list—invita- ticipated in the study for at least two years. tions to participate were therefore received by participants who had selected into the project management community Communications and whose organizations as a result could arguably be One of the participating organizations that provides the expected to demonstrate a higher degree of maturity than greatest understanding of how changes in maturity can the general population. Although the community within occur as organizational priorities change is illustrated in PMI has still received invitations through their chapters, this Figure 3. The graph provides an overview of the results for

68 AUGUST 2006 PROJECT MANAGEMENT JOURNAL 5.00

4.00

3.00

2.00

1.00 Program Project Project Plan Project Project Project Risk Technology Environment Organization Resource Contract Initiation Initiation Sched & Management Tracking Reporting Management Management Management Budget 1998 2.40 3.33 3.05 3.14 1.91 2.80 2.00 2.50 1.75 4.00 2.33 3.00 1999 3.63 3.34 3.41 3.19 2.24 2.65 1.87 2.50 2.28 3.39 2.52 2.46 2000 2.97 3.18 2.97 3.01 2.28 2.76 3.06 2.39 1.96 3.13 2.35 2.30 2001 2.46 3.14 3.02 3.08 2.35 2.67 3.20 2.29 1.95 3.30 2.41 2.49 2002 2.50 2.57 2.33 2.36 2.08 2.24 2.63 2.26 1.61 2.70 2.29 1.88 2003 1.55 1.73 1.57 1.70 1.25 1.57 1.75 1.53 1.85 2.21 1.79 1.31

Figure 3: Maturity results—Communications organization the organization for each year that they have participated in vide almost a leading indicator of potential problems; there the study. The organization is a significant player within the is a decline in maturity in almost every year for which the telecommunications field. They have participated in the study has been conducted. study each year since 1998, and provide an interesting case In discussing these results with the organization, what study in the evolution of process maturity. emerged was an interesting profile of why the observed Interestingly, the process capability areas show a fairly decline had occurred. The economics for telecommunica- consistent maturity of process from 1998 through 2001, tions companies has been a challenge for a number of years. with the project initiation, project planning, scheduling and The year 2002 saw an extensive number of layoffs, particu- budgeting, and project management capability areas scoring larly for senior staff within the organization, resulting in at or above level 3. For these four years, there was very little many senior project managers leaving the organization. variation in process results, except for a slight improvement Finally, the emphasis placed organizationally on effectively in project tracking after 1998, a slightly greater degree of managing by projects disappeared—instead, the philosophy maturity for project planning, scheduling and budgeting in shifted from ensuring a formal approach to one of “get it 1999 and a shift in the degree of maturity associated with done.” As a result, organizational participants strongly rec- program initiation. There was a larger variation in the orga- ognized the shift in maturity being reflected within their nizational capability areas, with there being a particular organizations results. improvement in risk management for 2000 and 2001 over the results for 1998 and 1999. Government What is particularly noteworthy in this organization’s The organization shown in Figure 4 had a similar profile in results is the significant decline in the overall demonstrated terms of their results, while not as drastic as that of the pre- process maturity that has occurred in the 2002 and 2003 vious organization. The organization is a municipal govern- results. The 2002 results show a decline of nearly one half a ment, with participation in the benchmarking study level in the process capability areas and approximately one coordinated by the PMO within the organization for each fourth of a level for the six organizational capability areas. year except 1999. The highest demonstrated maturity of the From 2002 through 2003, there was a subsequent drop of an organization again was reflected in 1998 and 2000; later additional one half a level in almost all capability areas. years demonstrated a corresponding reduction in overall Finally, the results for the organizational capability area pro- maturity. In 2001 through 2003, there was again a remark-

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 69 5.00

4.00

3.00

2.00

1.00 Program Project Project Plan Project Project Project Risk Technology Environment Organization Resource Contract Initiation Initiation Sched & Management Tracking Reporting Management Management Management Budget 1998 2.22 3.19 2.84 2.14 2.29 2.51 2.36 2.27 2.75 2.29 2.54 2.56 2000 2.92 3.92 2.63 2.19 2.05 2.26 2.32 1.88 2.28 2.29 2.36 2.11 2001 2.46 3.46 2.23 2.01 1.80 2.21 1.99 1.47 1.96 2.01 2.00 1.92 2002 2.29 3.38 2.24 2.08 1.90 2.04 2.42 1.91 1.95 2.23 2.04 2.16 2003 2.68 2.56 2.40 2.36 1.77 1.99 2.30 1.81 1.61 2.30 2.22 1.90

Figure 4: Maturity results—Municipal government organization

able consistency in terms of process, which was in part attrib- capability areas between the 2002 and 2003 studies, with utable to a common methodology being introduced in late the results varying by less than one fourth of a level in each 2000, but ironically this consistency appears to have come year. There has been a significant improvement in risk man- at the expense of greater maturity—the overall scores are agement, with an improvement of almost one half of a level lower than they were prior to the methodology being intro- over previous years. This result can reasonably be predicted duced. After 2001, there have been capability areas that given the emphasis the organization has placed on improv- have demonstrated improvement—in particular risk man- ing and formalizing its risk management approach since agement and technology, as well as the organizational capa- the presentation of the initial findings in 2002. bility. Interestingly, the net effect of the improvements from Although there is less variation in the results and a 2001 to 2003 had only been to restore these capability areas smaller number of time periods reflected than for the pre- to the levels they were at the outset. For the process capa- vious examples, this organization demonstrated the bility areas, the demonstrated maturity remained approxi- impacts that should appear from period to period where mately one half a level below that of the results for 1998. conscious improvement efforts are undertaken in response In reviewing the results with the organization, there to participating in the study. Given the emphasis on project was a high degree of face validity of the results being management within this organization, improvement in demonstrated within the benchmark study. Although overall maturity can be expected in future years. recent years have focused on improving the areas of risk management, technology, and organization, there has been Conclusions little coordinated focus on the processes since 2000, result- The previous section provides an assessment of organiza- ing in correspondingly lower results from 1998 and 1999. tional project management capability using descriptive sta- The decline has also been attributed to the champion of the tistics. Although the results that are conveyed are improvement effort retiring from the organization. individually interesting, it is helpful to explore how they contribute to our overall understanding of organizational Transportation project management. In this section, we summarize the The last organization is a much more straightforward exam- overall results, provide an assessment of the implications ple, and one that reinforces what is typically expected in and meaning that can be interpreted from these results, and comparing results of organizations year over year. As can be explore how a better understanding could be developed seen in Figure 5, there have been few changes in the process through subsequent research efforts.

70 AUGUST 2006 PROJECT MANAGEMENT JOURNAL 5.00

4.00

3.00

2.00

Figure 5: Maturity results—Transportation engineering organization

Observed Results • Since the initial decline reflected in the third year of The analysis produces some interesting observations, many the analysis, there have been few meaningful changes of which were not expected at the outset. A summary of the in assessed maturity—the overall maturity of organi- key observations within the study are as follows: zations since then has been relatively static. Given the • Overall, there has been minimal change within the stated focus that many organizations have on improv- demographics of the individual and organizational ing their project management capabilities, this raises participants with two specific exceptions. The overall significant questions as to the degree to which these composition of individual participants is relatively efforts are occurring or are having a demonstrable consistent, with the only change being an increase in impact. One possible explanation worthy of consider- the education of participants—particularly as demon- ation is that with the significant investment made in strated by either a certificate or a master’s degree in project management that occurred leading up to the project management or a related field. For organiza- year 2000 projects, the changes observed could also tions, there has been a greater emphasis on govern- suggest a subsequent abandonment of the capabilities ment organizations and proportionately less created during this period. participation from privately-held organizations. This • Within individual organizations that have participat- gradual shift in demographics does not explain the ed in multiple years, the results of improvement rapid shift in results being demonstrated, however. efforts as well as changes in strategy are strongly • Even with the shift in reflected maturity, interesting reflected in changes to assessed maturity. The impacts questions are raised about the profiles of both sets of of the changes reflected within the study are also organizations—those participating in 1998–1999 and being confirmed as valid in follow-up consultations those participating in 2000–2003. The demographic with the organizations. information suggest some changes that could have resulted: with the growth in governmental participa- Implications tion, there could have been a corresponding decline The observations and conclusions raised by this study pres- in overall assessed maturity. As well, the increase in ent some interesting implications for project management education suggest a greater awareness of the project in organizations. Even if the shift in maturity reflected in the management discipline, which could result in a more results is due to a change in those organizations choosing to accurate perspective of actual practices being provid- participate, rather than a real change in how project man- ed. Finally, the change in marketing away from the agement is practiced, it raises questions regarding the on- membership of PMI could indicate that subsequent going commitment of organizations to project respondents had less of a vested interest in ensuring management. The implication would be that those organi- assessed project management capability is high. zations that have been assessed as being more mature no

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 71 longer see value and relevance in benchmarking their capa- ing of the longitudinal changes that occur within a bilities. Although intuitively one would expect the opposite sufficiently large subset of participating organizations. to be true, the individual case studies provide two instances • Including a greater degree of follow-up with partici- that illustrate the tendency of organizations to make signif- pants to verify and validate the results being self- icant investments in project management only to subse- reported. This could include a definition of more quently abandon them. concrete and specific measures for project results, As well, the lack of any significant improvement in allowing a more granular understanding of the asso- maturity over the subsequent four years of the study pres- ciation between described project management ents implications regarding the improvement efforts being processes and delivered project results. The limitation conducted. On an anecdotal basis, many organizations pro- on the ability to gather this information, however, is fess to be making conscious efforts to improve their project the degree to which organizations track and are able management capabilities. While within individual organiza- to segment their project results to align with the tions there are some impacts reflected in assessed maturity defined measures. in response to specific improvement efforts, there does not • Including greater understanding of the dynamics appear to be a meaningful overall improvement as a com- occurring within organizations. Being able to build in munity in how project management is practiced. Similar to more of this context would help answer many of the the “productivity paradox” of IT investments, this raises questions underlying why changes were being questions as to the degree to which improvement efforts are observed that could not be answered in the current in fact being undertaken, and the effectiveness of these dataset. change programs in bringing about real improvements in • Looking at the outside influences on the organization, maturity and capability in terms of its economic environment, marketplace Overall, these results lead to a question about the role and competitive pressures. Understanding the market that project management has for organizations—whether it context will help to provide a better understanding of is viewed as a strategic enabler, core competency or simply how these factors also influence observed results. a fad whose time has come and gone. The most likely current answer, based upon both the stated commitment to Although the longitudinal analysis in this paper pro- improve organizations and the evidenced commitment of vides some valuable understanding of the context of proj- time and effort in participating in benchmarking exercises ects in organizations, the answers it provides are still such as this, is that project management is viewed as impor- fragmented and incomplete. By incorporating these changes tant by organizations but has not fully developed as an into future versions of the study, it is hoped that greater rel- organizational capability. As a result, its role as a strategic evance and insight will result that can constructively con- enabler or core competency is still in question, and is tribute to organizations better defining and realizing their something that organizations will still need to answer in the project management goals. In the meantime, this public coming years. benchmarking initiative continues as it originated, as a means to help organizations and project managers develop Opportunities for Further Refinement greater understanding of our complex world of work. The Although the descriptive statistics within this paper provides author wishes to think those individuals and organizations some useful insights into the changes that have occurred— that have contributed their time and effort to the bench- and not occurred—in how participant organizations marking study over the years, and who in doing so have approach developing and applying project management, it made the analysis in this paper possible. frankly raises as many if not more questions as it answers. Many of the questions arise from an inability to support fur- References ther analysis of underlying causes—this is in part a chal- Aiman-Smith, L., Goodrich, N. Roberts, D., & Scinta, J. lenge created by the benchmarking effort being largely (2005). Assessing your organization’s potential for value survey-based, without follow-on interviews, focus groups or innovation. Research Technology Management, 48(2), 37. case studies to provide additional context in understanding Biazzo, S., & Bernardi, G. (2003). Organisational self- the choices and influences of organizations. assessment options: A classification and a conceptual map Some specific opportunities for further improvement of for SMEs. The International Journal of Quality & Reliability this analysis include: Management, 20(8/9), 881. • Providing a more rigorous and comprehensive statis- Conlon, E. J., & Short, L. O. (1984). Survey feedback as tical analysis of the underlying data. Although this a large-scale change device: An empirical examination. paper provides a descriptive analysis of the available Group & Organization Studies (pre-1986), 9(3), 399. statistics, a more in-depth statistical analysis could Cooke-Davies, T. J., & Arzymanow, A. (2003). The provide additional information to support or refute maturity of project management in different industries: the inferences being drawn here. An investigation into variations between project manage- • Encouraging repeat participation from a greater num- ment models. International Journal of Project Management, ber of organizations, providing a greater understand- 21(6), 471.

72 AUGUST 2006 PROJECT MANAGEMENT JOURNAL De Vries, H., & Margaret, J. (2003). The development management maturity. Project Management Journal, of a model to assess the strategic management capability of 31(1), 32. small- and medium-size businesses. Journal of American Jachimowicz, V. A. (2003). Project management matu- Academy of Business, Cambridge, 3(1/2), 85. rity model. Project Management Journal, 34(1), 55. Deming, W. E. (1993). The new economics: for industry, Jugdev, K., & Thomas, J. (2002). Project management government and education. Cambridge: Massachusetts maturity models: The silver bullets of competitive advan- Institute of Technology. tage? Project Management Journal, 33(4), 4. Duncan, W. J., Ginter, P. M., & Swayne, L. E. (1998). Karapetrovic, S., & Willborn, W. (2001). Audit and self- Competitive advantage and internal organizational assess- assessment in quality management: Comparison and comment. The Academy of Management Executive, 12(3), 6. patibility. Managerial Auditing Journal, 16(6), 366. Garrett, G. A., & Rendon, R. G. (2005). Managing con- Kwak, Y. H., & Ibbs, C. W. (2000). Calculating project tracts in turbulent times: The contract management matu- management’s return on investment. Project Management rity model. Contract Management, 45(9), 48. Journal, 31(2), 38. Hellsten, U., & Wiklund, P. S. (1999). Self-assessment Mullaly, M. E. (2004). Executive Summary of the 2003 as a facilitator for learning. Quality Congress. ASQ’s… Organizational Project Management Baseline Study. Annual Quality Congress Proceedings, 423. Edmonton, Interthink Consulting Incorporated. Hillson, D. (2003). Assessing organisational project Nielsen, W. R., & Kimberly, J. R. (1976). Designing management capability. Journal of Facilities Management, assessment strategies for organization development. 2(3), 298. Human Resource Management (pre-1986), 15(1), 32. Holland, C. P., & Light, B. (2001). A stage maturity Sawaya, N., & Trapanese, P. (2004). Measuring project model for enterprise resource planning systems use. management maturity. SDM, 34(1), 44. Database for Advances in Information Systems, 32(2), 34. Skulmoski, G. (2001). Project maturity and compe- Humphreys, W. (1992). Introduction to software process tence interface. Cost Engineering, 43(6), 11. improvement. Pittsburgh, PA: Software Engineering Tuttle, T. C., & Romanowski, J. J. (1985). Assessing per- Institute, Carnegie Mellon University. formance and productivity in white-collar organizations. Ibbs, C. W., & Kwak, Y. H. (2000). Assessing project National Productivity Review, 4(3), 211.

MARK MULLALY has more than 20 years of project management experience, working in a variety of industries, including information technology, communications, utilities, insurance, finance, engineering, construction, high tech and the arts. As president of Interthink Consulting Incorporated, a company based in Edmonton, Alberta, Canada, Mark has been a pioneer in the development of organizational project management capabilities. Interthink is renowned for its Organizational Project Management Baseline Study and PM:RoadMap organizational assessment tool, both of which are based on Interthink’s project management maturity model. Current areas of work include the strategic alignment of projects and the impacts of personality on project manager development and competency. Mark is a popular and engaging speaker who has been invited to address conferences across North America. His widely read column, Project Management in Practice, appears monthly on gantthead.com, and he is an instructor in project management with the School of Business at the University of Alberta. He is also currently pursuing a doctoral degree at University of Technology, Sydney. Mark is the past president of the Northern Alberta Chapter of the Project Management Institute.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 73 DEVELOPING ORGANIZATIONAL PROJECT MANAGEMENT CAPABILITY: THEORY AND PRACTICE

LYNN CRAWFORD, ESC Lille, France and University of Technology, Sydney, Australia

Introduction ABSTRACT roject management as a field of practice has been brought into being through the conversations, writing, and collaborative activities of practitioners, con- This paper traces the evolution of con- P sultants and academics with a shared interest in dealing with phenomena ceptions of project management from the use of tools and techniques on stand- that are perceived to have similar characteristics and challenges (Parker, 1992). alone projects to the conceptualization As project management evolves in this way as a field of practice, there is often of project management as an organiza- an interesting tension between practitioners and academic researchers in project tional capability. Working from the prem- management, with the practitioners claiming that the discourse in the field is too ise that project management is a socially theoretical while the academics claim that it lacks theoretical foundations (Betts & constructed field of practice that has developed through the conversations Lansley, 1995; Koskela & Howell, 2002; Shenhar, 1998). Although, in a strict and deliberate efforts of practitioners, sense, the academics are correct, the field is rich in espoused theories of practi- principles of discourse analysis are used tioners (Argyris, 1995; Argyris & Schon, 1977). Project management is construct- as a framework for studying the extent to ed by the actions and interactions of practitioners, consultants, and which practice reflects the espoused the- academics/researchers through their use of language, communication of beliefs, ories of organizational project management capability development. The and interaction in social situations (van Dijk, 1997), and as represented in written actuality of practice is represented by and spoken language, cultural artifacts, and visual representations (Hardy, 2001). periodic reports over a five-year period Discourse analysis, the systematic study of texts, provides a useful framework for by the “owners” of project management gaining an enhanced understanding of the nature and evolution of project man- in an organization with an expressed agement theory and practice. This paper uses the principles of discourse analysis commitment to development of organizational project management capability as a framework for study of the relationship between espoused theory and practice and is analysed with reference to the concerning organizational project management capability. In this analysis, the related espoused theories of practition- actuality of practice is represented by periodic reports over a five-year period by the ers as represented in the project man- “owners” of project management in an organization with an expressed commit- agement literature, including bodies of ment to development of organizational project management capability. The prac- knowledge, standards, and guides. titioner experience is presented against a backdrop of the evolution of conceptions Keywords: organizational project man- of project management from the use of tools and techniques on stand-alone proj- agement capability; theory and practice ects to the conceptualization of project management as an organizational capability. This evolution is presented through discussion of the espoused theories of ©2006 by the Project Management Institute organizational project management capability as represented in the project man- Vol. 36, No. 3, 74-97, ISSN 8756-9728/03 agement literature, including bodies of knowledge, standards, and guides.

Background Discourse analysis provides a way of studying phenomena through interactions and the use of language in order to gain new insights and understanding. From this perspective, discourses are considered constructive in that they “bring reality

74 AUGUST 2006 PROJECT MANAGEMENT JOURNAL into being by making social relations others. A further and related consider- The mid-1990s were a crucial and material objects meaningful” ation is the need for reflexivity (Hardy, point in the development of project (Hardy, 2001). Discourse and its effects 2001), whereby the researchers reflect management standards and related are therefore constantly shaping and on their role and the influence that certification programs. Indicative of being shaped by their context and by their own interests and choices will the conception of project management other discourses. This paper looks at have on interpretation and outcome. at this stage in its development, all of two related discourses. The discourse This discussion on discourses is the emerging standards focused on chosen as representative of espoused organized into three sections: stand-alone projects and individual theory (Discourse 1) is the discourse • Section 1: Evolution of Concepts project management practitioners. The evidenced and constructed by talk and of Project Management provides Project Management Institute issued A texts through the project management an introduction to Discourse 1 Guide to the Project Management Body of literature and in the form of standards through an overview of the histor- Knowledge in 1996, and in the same and guides. The discourse representing ical and social setting for both year the Association for Project practice (Discourse 2) is that of owners espoused theory and practice. Management in the U.K. issued the of project management in one organi- • Section 2: Conceptualization of Third Edition of its Body of Knowledge. zation over a four-year period. This dis- Organizational Project Management The Australian and United Kingdom course operates within, and is Capability gives brief coverage of governments endorsed performance- influenced by, the wider context of the relevant espoused theories of based competency standards for project Discourse 1. practitioners. managers in 1996 and 1997, respec- In undertaking an analysis of this • Section 3: Practitioner Experience tively. The British Standards Board also nature, important elements are the in Developing Organizational issued their Guide to Project Management choice of talk and text, and the loca- Project Management Capability in 1996. The International Project tion of discourses in their historical describes the specific texts and con- Management Association issued their and social context. According to van text of the discourse relating to the IPMA Competence Baseline in 1998. Dijk (1997, p. 4), we may need: actuality of organizational project From the mid-1990s onward, • Theoretical notions that define management capability develop- interest in project management grew the beginning or the end of text ment (Discourse 2). progressively stronger, with a move and talk towards the concept of project man- • Their unity or coherence Evolution of Concepts of Project agement as an organizational capabili- • Inter-textual relations between Management ty, fuelled by a series of articles in different discourses Project management can be seen as a PMNetwork by Paul Dinsmore (1996a, • Intentions of speakers or writers socially constructed field of practice 1996b, 1996c, 1996d) who has consis- • Settings, time, place, and other that has developed from tools and tently acted as a chronicler of project aspects of the communicative techniques designed to support the management practice. In this period context. management of major projects, from also, an interest in benchmarking of the conversations of practitioners and corporate project management prac- It is important to know something from their deliberate efforts to define a tices emerged. Two notable initiatives about who is talking or writing, as field of practice through definition of a were the PMI-supported Fortune 500 what is said is the negotiation of distinct body of knowledge and associ- Project Management Benchmarking meaning and will be affected by the ated standards. Fundamental to this is Forum, which was formed in the mid- interests and power relations of the recognition of projects as phenomena 1990s, and the Human Systems various actors and groups of actors. For with shared characteristics. Knowledge Network, which started example, practitioners in organizations The first signs of project manage- collecting organizational project man- may express desire for a common lan- ment as a distinct field of practice agement practice data and facilitating guage with shared meanings for use in were the network analysis and plan- knowledge sharing between corporate relation to projects. Consultants and ning techniques, like PERT and CPM, owners of project management in project management associations, that emerged in the 1950s for use on 1993. Both initiatives have contributed however, may have a commercial or major projects in construction, engi- to the development of the concept of proprietary interest in encouraging neering, defense, and aerospace indus- organizational project management specific or more diverse terminology tries (Kerzner, 1979; Morris, 1994; capability through publication and and meanings. In organizations, the Stretton, 1994a). Users of these tools conference presentations. Meanwhile, specific interests and agendas of the and techniques recognized shared the majority of key project manage- actors will influence what they consid- interests leading to the formation of ment professional associations have er important and the ways in which project management professional broadened their attention from facili- they present it. What they say may be associations in the late 1960s, initially tating the knowledge sharing and pro- influenced by issues of ego or concern to facilitate knowledge sharing fessional development of individual that what they say may be reported by between practitioners. practitioner members to engaging and

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 75 addressing the needs of what they term practitioner- and association-led initia- ment as an organizational capability, “corporates” either as a new class of tives for development of standards and has been the development and promo- membership or in other ways as key guides that structure understanding of tion by the U.K. government of a project stakeholders. organizational project management management methodology, PRINCE2, This change from focus on the capability. A key issue is recognition initially designed for use on IT projects individual project and practitioner to that in this broader application, proj- but further developed for wider applica- project management as an organiza- ects in organizations are rarely isolated tion. Although the standards and guides tional capability reflects the wider from environments that organizations for management of individual projects adoption of project management and a must balance the re-sourcing of portfo- focused on project-related practices change in the nature of the concerns lios of projects, and that more than relating to time, cost, quality, risk, and conversations of this broader one project may be responsible for the human resources, communication, and group of practitioners. When the project delivery of the same strategic goal or procurement, the shift toward project management associations first devel- set of outcomes or desired benefits. management as an organizational capa- oped, the conversations between mem- bility has been accompanied by interest bers involved senior project managers Conceptualization of Organizational in benefits management and gover- of large and often high-profile projects. Project Management Capability nance which are featured in both As the disciplines developed on these The previous section has given a brief PRINCE2 and Managing Successful major projects have been adopted “to overview of the historical and social Programs (MSP), developed and pro- cope with the management of employees setting of the field of project manage- moted by the U.K. Office of involved with irregular assignments and to ment. It focused on the evolution of Government Commerce, ostensibly to apply a structure to complex and discontin- the concept and context of project help public sector organizations to uous undertakings” (Hodgson, 2004, p. management through the interactions improve their efficiency, gain better 82) in finance and other sectors, the of practitioners. The current section value for money from their commercial actors and their context have changed. will focus on the evolution of the con- activities, and deliver more successful There are now many conversations tak- cept of organizational project manage- programs and projects. ing place at many different levels. The ment capability (OPMC) as a specific Development of the Project shift can be seen in the membership discourse within the wider field of Management Institute’s Organizational and participation of the project man- project management. The intent is to Project Management Maturity Model agement associations, which are now provide the context for the following (OPM3®) commenced in 1998 and was dominated by consultants, trainers examination of a specific organization- released in 2003 (Project Management and relatively junior project managers al discourse as a basis for comparing Institute, 2003). During this time, it and team members. Staff and leader- espoused theories, represented by generated its own discourse with several ship of the associations conduct con- Discourse 1 with theories in use or hundred volunteers taking part in the versations with the senior practice (Discourse 2), and to test discussions, the talk, and the text sur- management of “corporates,” who whether development of OPMC in rounding its development. Although may have no direct experience in man- practice reflects the espoused theories the content of OPM3 is not widely agement of projects. The managers of as presented in the literature and stan- known beyond those who were major projects whose shared experi- dards for practice. involved in its development, and ence and interactions led the develop- Extension of the focus of project because it is potentially too diverse ment of the field until the early to management beyond the individual (with more than 600 “best practices,” mid-1990s now tend only to appear as project to encompass multiple projects, more than 3,000 “capabilities,” and the occasional invited keynote speaker programs, portfolios, and enter- more than 4,000 relationships between at a conference. With a change in the prisewide approaches has changed the capabilities [Cooke-Davies, 2004]) to actors and their context, the nature of context, the actors, and the nature of have clear impact on the construction the discourse has changed. conversations between them. The and conceptualization of practice, it The desire of senior practitioners changing nature of the discourse is has already had a pervasive influence to share and codify their experience in reflected in the commencement of on the discourse by institutionalizing management of major projects has development in 2005, by the Project the notion of project management been replaced by the desire of relative- Management Institute, of standards and maturity. As early as 1998, the PMI ly junior practitioners for training and guides for Program and Portfolio Standards Committee established a certification for career advancement Management and the development, by standards project that was initially and the desire of senior managers for the Association for Project Management conceptualized as a guide to creating guidance in development of organiza- in the United Kingdom, of A Guide to organizational environments to sup- tional capability, one aspect of which Governance of Project Management port management of projects1. Both is the project management compe- (Association for Project Management, Graham and Englund (1997) and tence of their personnel. This change 2004). Another strong voice in the Dinsmore (1999), who was a member in focus has been accompanied by conceptualization of project manage- of the PMI Standards Committee,

76 AUGUST 2006 PROJECT MANAGEMENT JOURNAL contributed to development of this models (the OPM3 team examined ature. Associated with this is strategic concept. However, early development of more than 30 extant models (Cooke- alignment of projects and programs OPM3 was influenced by the discourse Davies et al., 2001), and other with organizational aims. Further, in software engineering around capabil- approaches to organizational project increasing application of project man- ity maturity (Humphrey & Sweet, 1987; management capability are, in the lit- agement to internal projects—particu- Paulk, Weber, Curtis, & Chrissis, 1995), erature, often left unstated. When larly in business—changes and, in the and the consequent emergence of a looked at they have strong similarities financial and government sectors, has number of project management maturi- and some differences. However, while raised interest in benefits management ty models in the mid-1990s interpretations may differ across and governance, both of which have (Cooke-Davies, 2004; Cooke-Davies, Sc industries, application areas and also been highlighted by the wider dis- hlichter, & Bredillet, 2001; Pennypacker regions, the concept of organizational course on corporate governance. & Grant, 2003), so it is not surprising project management capability and of Leadership, performance manage- that the initial idea was re-formulated as maturity of that capability has become ment, and top management support, an organizational maturity model. a widely accepted feature of the including the role of the project/execu- In Europe, the concept of organiza- discourse. tive sponsor have also attracted tional project management maturity Another strong emergent theme in increasing interest in recent years. has been reinforced by the work of organizational project management is Appendix A of this paper presents top- Roland Gareis who has been a leader in the project or program management ics and themes that are representative promotion of the concept of manage- office (PMO), an organizational entity of the espoused theories of project ment by projects (Gareis, 1992a) rather established to provide coordination or management in general, and for orga- than the traditional concern with man- support for management of a number nizational project management capa- agement of projects. In the early 1990s of projects or programs. Although it is bility in particular. Gareis talked of project-oriented com- generally agreed that one size does not panies performing “simultaneously fit all, there is some consistency in the Practitioner Experience in Developing small and large projects, internal and types of functions provided, as found Organizational Project Management external projects, and unique and in studies reported by Crawford Capability repetitive projects to cope with new (2004b), Hobbs and Aubry (2005), The previous section has presented the challenges and potential from a and Dai and Wells (2004). See Table 1. discursive context, which might rea- dynamic business environment” Examination of a number of stud- sonably be expected to influence peo- (Gareis, 1992b). He also talked about ies of trends and topic coverage in the ple in organizations as they discuss the need to support the performance of project management journals (Betts & and take action to develop corporate projects with adequate strategies, struc- Lansley, 1995; Crawford, Pollack, & project management capability. In tak- tures, and cultures. England, 2006; Kloppenborg & Opfer, ing action, they will be drawing ideas Through the discussions, confer- 2000; Morris, 2000; Morris, Patel, & from a range of sources and may influence presentations, and papers of con- Wearne, 2000; Themistocleous & enced by this wider discourse in for- sultants, academics and practitioners, Wearne, 2000; Urli & Urli, 2000; Zobel mulating and articulating plans project management as an organiza- & Wearne, 2000), the content of a and actions. tional capability has become an number of the maturity models and Alvesson and Karreman (2000) important focus for discourse in the other publications relating to aspects stated that “with a lot of discourse talk, field. Strongly associated with this are of organizational project management it is sometimes rather unclear what ideas of assessment and development capability, reveals common themes. ‘discourse’ refers to” (p. 1140). As with in terms of capability maturity. As Clearly, the PMBOK® Guide both any study, there are multiple discours- Cooke-Davies suggested, maturity reflects and has had a pervasive influ- es. This study is specifically comparing models “seek to do for organizations ence on the rhetoric of both manage- two identified discourses—the concep- seeking to implement strategy through ment of, and by, projects, as tualization of organizational project projects what ‘bodies of knowledge’ integration, time, cost, quality, human management capability as represented have done for individual practitioners resources, communications, risk, and in the standards, journal articles, con- seeking to improve their ability to procurement appear consistently in ference papers, and presentations pro- manage projects” (Cooke-Davies, both the general project literature and, duced by the project management 2004). Interestingly, while there is in one form or another, in many of the community (Discourse 1); and the much written about maturity models, maturity models. From an organiza- practice of organizational project man- the focus is not so much on the con- tional perspective, they are generally agement capability development as tent as on the concept of maturity associated not only with project represented in the reports of practice in itself. Although the concept of maturi- processes, but in some cases at pro- one organization over a four-year peri- ty is generally accepted and much dis- gram or portfolio level as well. od (Discourse 2). There are other discussed, the aspects of capability that Program and portfolio manage- courses that influence the actors in the are assessed in the various maturity ment are emergent themes in the liter- organization that is the focus of study

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 77 Hobbs & Aubry, 2005Dai & Wells, 2004 Crawford, 2004 Monitoring and controlling project Providing project administrative support Planning and control support performance reporting

Development of PM competencies and Developing and maintaining PM PM methodology and standards methodologies standards and methods PM tools Providing PM consulting and mentoring PM competency and career development Providing or arranging PM training

Strategic management Linking projects to strategic goals

Multiproject management Providing human resources/staffing Resource management assistance

Organizational learning Developing and maintaining project Audit/review historical archives Lessons learned and continuous Improvement Communications and PM community

Purchasing and contract administration

Table 1: Functions of a project or program management office and that should be acknowledged. spread discourse—in this case, the relationship between many of the par- Although discussed and referred to as wider discourse on organizational ticipants has developed beyond apparently coherent entities, organiza- project management capability— acquaintance, to friendship, and there tions are themselves an emergent “shapes both how to talk about subject is considerable background knowledge property of discourse. Within each matter and the meanings that we about other organizations and the his- organization, there will be a number develop about it” (p. 1138). The focus tory of their improvement initiatives. of competing discourses; for instance, of concern is a perceived general ten- Regular attendees are familiar with the between functional divisions. dency relative to the research question process and with one another, and this Organizational representatives whose which is the extent to which the reality creates an environment in which new voices contribute to the specific dis- of practice reacts to, reflects, and/or attendees are quickly influenced by a course under study (Discourse 2) will influences the espoused theories of culture of open sharing of knowledge be influenced by the talk and text of organizational project management and experience. All organizations rep- their organizational environment and, capability and its development. resented can be assumed to have a as they may come from different disci- shared and active interest in improving plines—such as engineering, IT, Data: Text and Context their organizational project manage- human resources, accounting, or busi- The text that has been selected as the ment capability as they have made a ness administration—their world view basis for this analysis of organizational financial commitment to membership. and language may be influenced by the project management capability devel- In fact, the networks have the charac- discourse in these fields. Each of these opment in practice is a transcription of teristics of an effective community of overlapping and intersecting discours- reports made several times a year by practice (Hildreth, 2000; Wenger & es potentially shape and are shaped by organizational representatives attend- Snyder, 2000). one another. The boundaries between ing knowledge-sharing workshops as One of the characteristics of an the various discourses are permeable, members of the human systems proj- effective community of practice is a and the definition of a particular “dis- ect management knowledge and common language. For members of the course” is therefore a theoretical dis- benchmarking networks. Membership human systems networks, this is pro- tinction. As van Dijk (1997) said, the is entered into on an annual basis so vided by the corporate practices ques- definition of the beginning or end of that there is continuity in participation tionnaire (CPQ), a tool used for text and talk is a theoretical notion. in workshops, a process that facilitates assessing, auditing, and benchmarking ongoing conversations. Although there of corporate project management prac- Approach are usually a number of attendees at tices. Each member uses this and other The approach taken can be described workshops that have not been there assessment tools to baseline their cor- as a long range/determination position before, there are always those who porate project management capability, (Alvesson & Karreman, 2000) in which have attended on more than one occa- to develop a plan for improvement and it is assumed that dominant and wide- sion, often over a number of years. The then, through further assessments, to

78 AUGUST 2006 PROJECT MANAGEMENT JOURNAL demonstrate results. Although not con- taken from the talk of member repre- text. These codes related both to charac- ceived as a “maturity model,” as sentatives that forms the basis for the teristics of the text (e.g., nature of language described earlier in this paper, the CPQ following analysis. used) and the context (e.g., business can be considered in this light, and change/restructuring) (Appendix A). results can be mapped to CMMI Methodology Where text related to more than one topic (Chrissis, Konrad, & Shrum, 2003), The transcripts of verbal reports from or theme, multiple codes were applied. OPM3 and other maturity models. The one of the member organizations have For instance, if cost was mentioned in the CPQ has been shaped by, and shapes, been selected to represent Discourse 2, context of organizationwide project man- the discourse of network members. It the development of organizational agement processes or methodology, the was first developed in 1994, by mem- project management capability in text would be coded both for “Cost bers of the network, and has subse- practice. The organizational text was Management” (6) and “PM Policies, quently been continuously used and chosen at random from more than 20 Processes, Methodology & Tools” (58). developed by them. As the instrument possible sets of text. Selection of a sin- is used by all member organizations as gle case study allows in-depth analysis Context a basis for their capability improve- of the text in order to examine the To maintain confidentiality, the organ- ment programs, its language is familiar extent to which the theory in use or ization whose report transcriptions to all participants. It provides a com- actuality constructed by the discourse have been selected for analysis will be mon language that is widely used by (Discourse 1) reflects the espoused referred to as The Organization. network members and therefore forms theories as documented in the litera- Reports cover from 2000 to 2004. At an important aspect of their discourse. ture and standards (Discourse 1). A the start of this period, the IT division, It both constructs and is constructed by significant feature of the chosen dis- tasked with improving their perform- the talk and text of network members. course (2) is that the text does not just ance in management of projects, Workshops of the network are represent one point in time, but is rep- joined the Network to enable them to designed to provide members with an resentative of a four-year period. So, by establish a baseline for performance opportunity to address specific issues looking at one organization, it is pos- using the Network’s assessment tools, that are of concern to them, to learn sible to examine the characteristics and to identify strengths and weaknesses from other members, and to share and development of the discourse over and to develop an improvement plan. create knowledge relating to organiza- time as it interacts with and affects Within a year of joining the Network, tional project management capability. both global and local context (van the IT capability of The Organization The members select the themes and Dijk, 1979, p. 19). was outsourced. Following a presenta- topics for the workshop in a process The transcripts of reports have tion to representatives of a number of that relies upon discussion and negoti- been coded by two researchers and divisions, The Organization decided to ation. Although the format of work- analyzed using proprietary text analy- continue its membership of the shops varies, there has been one sis software. Both a priori and emer- Network. Initially, this was driven by consistent element—a report by mem- gent codes were used. The a priori the internal audit function of The ber representatives, usually at the start codes were chosen as representative of Organization, which saw a strong link- of each workshop, on current concerns the espoused theories of project manage between the performance of proj- and project management improve- agers and researchers as presented in ects, particularly related to change, and ment activities and achievements since the project management literature and the financial performance of the the last workshop. As member repre- standards. These codes were based on organization, and were therefore very sentatives verbally deliver their brief 48 topics identified from review of supportive of organizational improve- reports, they are directly transcribed topic coverage in project management ment to manage projects. onto computer, following the wording journal articles (Themistocleous & Throughout this period, The presented as closely as possible, and Wearne, 2000) and project manage- Organization has had over 20,000 this is projected to enable on-the-spot ment standards and guides (Crawford, employees, and a number of local and corrections or amendments by mem- 2004a; Crawford et al., 2006). To this globally distributed subsidiary compa- bers. Only on rare occasions have were added a number of terms repre- nies. It is an extremely complex organ- members requested that any informa- sentative of organizational project ization and the part directly tion be suppressed, sometimes because management as outlined in the con- participating in the Network has a of information embargoes affecting ceptualization of organizational proj- number of divisions or business units. public companies, and sometimes ect management capability section of As often happens, “ownership” of because the information shared could this paper. A small number of emer- Network membership was required to be of a commercially sensitive nature. gent codes represent recurring themes be held by one division, and this These transcribed reports are made in the transcripts of reports that were changed from time to time during the available to all members by e-mail and not clearly covered by the a priori period of membership. As Network via a member intranet as part of a codes and which became evident to membership requires the nomination workshop review report. It is the text the researchers as they reviewed the of one or two key representatives, these

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 79 representatives and the guests they topics/themes are with organizational in The Organization. It is particularly bring with them to Network meetings project management capability rather interesting that the majority of the will change from time to time. Other than with individual projects. themes identified in the text did relate reasons for change in the individuals Topics and themes mentioned but to organizational project management participating are staff changes includ- with lower frequency and not in each capability and that terms more directly ing resignations, retirements, promo- of the four years were as follows in related to management of individual tions, and relocations. However, there Table 2. projects, specifically time/schedule is generally a reasonable degree of con- Other topics and themes listed in and cost did not feature strongly even sistency, and this was the case with The Appendix A were not evident in the in the context of organizationwide Organization. The main changes were text. This initial analysis confirmed application. The following discussion from the IT division, to the internal that the reality of practice reflects the will deal with each of the popular audit function, and then to representa- espoused theories of organizational themes, and will include treatment of tives of the corporate program office project management capability and its changes in emphasis of each of these when it was formed in 2003. development. The only topics and themes over the period of study. themes not mentioned in practice are Analysis maturity and ethics/rules of conduct. Organizational Project Management The relationship between the reality of Improvement programs were generally Capability practice (Discourse 2) and the espoused implied and improvement in a general PMO/Support Office theories of organizational project man- sense mentioned occasionally, but no This was the most important and agement capability and its development coherent approach to capability recurring theme in the text, appearing (Discourse 1, Appendix A) was first improvement was evident in the text. in every year of the study period, but analyzed by comparing the recurrence Analysis of trends over the period of peaking in 2003 when an enterprise of themes from Discourse 1 in the study will be covered in the following, program office (EPO) was established. transcripts of reports from The more detailed discussion. The concept was first mentioned in Organization (Discourse 2). The focus 2000 in the context of a “pilot scheme of their reports was current concerns Discussion where the project office works with the and project management improve- A preliminary analysis of key themes project manager before startup of a ment activities and achievements at in the text indicates a relationship project as a way of improving project various times over a four-year period. between the discussion of organiza- performance and delivery.” The voice The most popular themes or top- tional project management in the liter- at this time was that of the IT division ics for The Organization, over a four- ature and standards (Discourse 1) prior to the outsourcing of this func- year period, were: representing espoused theories and the tion. In 2001, there was reference to • PMO/support office expressed concerns and project man- “each division setting up their own • Reporting agement improvement activities and PSO or each providing similar servic- • Project management policies, achievements of “owner” of organiza- es” and the impact of the Network dis- processes, methodology, and tools tional project management capability course was evidenced by a report that • Benefits management • Project management compe- Mentioned More Than Once Mentioned Only Once tence and career development • Governance • Project monitoring control • Top management support • IT/software • Sponsorship • Resource management • Business change • Risk management • Requirements management • Management by projects • Quality management • Quality management • Organizational learning. • Program management • Estimating • Community • Culture Of these themes, only PMO/support • Strategic alignment • Cost management office; reporting; project management • Project initiation/startup • Contractors policies, processes, methodology and • Outsourcing tools; benefits management; and project • Project Closeout/finalization management competence and career • Benchmarking development were mentioned in every • Information/communication year. Emergent themes relating to context Management are marked in bold, indicating that busi- • Stakeholder/relationship ness change and IT/software were signif- Management icant issues for The Organization at one Note: Items marked in bold are emergent themes. Items in italics relate to organizational project time or another during the period of management rather than the managemant of individual projects. study. All of the most popular Table 2: Lower frequency topics and themes

80 AUGUST 2006 PROJECT MANAGEMENT JOURNAL “a couple of the business units that ber of committed and interested indi- reviewing methodologies that were attended the last workshop have used viduals. Messages became mixed— “different across various areas.” They that input in setting up their own varying from a recap of the formation were “aiming to get people to use the PSOs.” In 2002, the discussion contin- of the EPO in 2003, to expressions of same methodologies,” but by 2003, ued to refer to divisional project sup- concern, on more than one occasion, they were meeting resistance and port functions. In 2003, an Enterprise that it was a “diverse organization” finding that “some of the processes Program Office was established based and that there was “no clear central appear too cumbersome to project on a written recommendation to man- program office.” By the end of 2004, managers.” However, there were agement by the internal audit function the role of the EPO had been refo- other indications that the pressure to “to improve commercial outcomes of cused on support of organizational perform was increasing as well as a projects across [The Organization].” change initiatives and was restated as concern that project managers were Near the end of 2003, the repre- that of consolidating “reporting (cost “just getting projects done quickly so sentative of The Organization (includ- and benefits) and monitoring of KPIs” that benefits can be realized—often ing representatives of internal audit of the change initiatives. It was stated without due process.” and one or two divisions) reported that the EPO, now apparently a some- By 2004, they had initiated a that there were “good early signs.” what different entity than had been set specific project to establish a “stan- Indicative of the different views that up at the start of 2003, was “charged dard methodology through the exist even within one organization, with increasing execution capability” entire life cycle of projects” across there were varied perceptions of the of the organization. the organization and the earlier con- expectations, roles, and function of the cern about “cumbersome” processes EPO. While the divisions were looking Reporting was reflected in a desire for “mini- to it for “solutions and support,” it was Discussion of reporting peaked in mum governance and reporting.” A perceived by some as “more financial- 2004. Up until this time, it was prima- sense of tension between desire for ly and reporting driven,” yet earlier it rily discussed in passing as one of the corporate control and standardiza- was claimed that the purpose was to roles of the EPO, although there was tion and corporate pressure for per- “go well beyond reporting,” operating mention of a “home grown project formance, allied with project in “a more low profile way, responding tracking system… plus Web-based proj- management reluctance to follow as required to the circumstances that ect reporting system from the central process, emerges from the text. arise in order to keep the wheels turn- server,” which would enable a central ing.” Others described the intended registry of projects. Little more was said Benefits Management role of the EPO as “looking at a high specifically about reporting until 2004. Benefits are referred to in each year level vision for projects” including the Early in the year, they were “looking at of the study period but with increas- “financial side from business plan and the reporting of small projects—which ing emphasis in 2003 and 2004. business case to business implementa- together are to deliver significant ben- Benefits tracking is first mentioned tion review and getting this across to efit and add up to a considerable cost” in 2000. In 2001, there is mention of the project managers who often come with “concerns about data integrity in the rollout of a tool for benefits real- from an operational background.” A reporting.” These and other concerns ization and reporting “with interest- concern was expressed that many proj- appear to have influenced a move to ing feedback.” Benefits are discussed ects were not defined as such, were implementation of software to pro- in association with “definition, funded from operational budgets, and vide integrated project reporting development and monitoring of did not have “visibility.” Also near the across the organization. KPIs” and value of project benefits is end of 2003, there was mention of an related to total project “spend.” This “intranet site that sets some high level PM Policies, Processes, Methodology, and Tools discussion has a board-level dimen- policies in terms of projects—business Discussion of this topic was fairly sion, reflected in the statement plan, business case, benefits realiza- even throughout the period of study, (2002) that the “board has realized tion, status reporting formats and but peaked in 2004, as did the vol- how much is spent on projects so are planning guidelines.” ume of discussion in general. In focusing on linking to benefits” with Although the EPO was only estab- 2001, although senior management attempts to “express project benefits lished in early 2003, there was already had “given authority to achieve con- in relation to P & L on the balance mention in that year of “difficulties in sistency in PM processes,” the orga- sheet.” The corporate project man- terms of management buy-in to sup- nizational representatives reported a agement representatives were con- port continuity,” and also of the set-up “challenge to apply a project/pro- cerned, however, that senior of divisional support offices. In 2004, gram discipline across the group” as management “doesn’t recognize the participation in the Network they had “pockets of project man- what is required to improve scoping changed. Internal audit was no longer agers with different processes and and benefits reporting.” Throughout participating in the network, and par- governance.” This was further 2003 and 2004, there is increasing ticipation was largely by a small num- addressed in 2002, in the context of reference to “benefits realization.”

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 81 PM Competence and Career Development in place prior to this time, it seems that dard methodology are overtaken by This topic has a number of subsidiary the term started to become popular to focus on the identification of software themes, including the following: cover a range of activities that might that will provide enterprisewide “proj- • Training well have been referred to in a different ect and program perspectives.” In • Mentoring way previously. In 2001, The selecting the software, they expressed a • Human factors Organization representatives recognize desire for “full integration of reporting • Competence that there are different approaches to and other functionality” and confi- • Career development governance throughout the organiza- dence that the supplier was a “market • Accreditation. tion and in 2002 identify a “split leader in development of this kind of between project governance and busi- project.” It was clearly also important Training and competence are the ness-as-usual governance.” Although it that the software providers were “very predominant subthemes. As early as is seen as important, there are three ref- amenable to making changes” and 2000, there is an expressed aim to erences to a desire for “minimum gov- “easy to deal with.” achieve professional accreditation of ernance,” which appear related to the project managers across the organiza- expressed concerns not to burden the Management by Projects tion and there is reference in 2002 and project management community with An organization-wide approach to 2003 to application of project/program “cumbersome” processes, especially if project management is inherent in discipline and need for professional they may stand in the way of “delivery.” much of the general reporting from the training of project managers, many of Governance potentially represents representatives of The Organization, as whom are thrown into the position. One an instance of the influence of practice already discussed above. In each reason given for concern with compe- (Discourse 2) on Discourse 1, as it is a instance, when they refer explicitly to tence of project managers was that they relatively new theme in the project projects as a way of doing business, it had “a couple of projects that could management literature. While gover- is associated with direct or implied ref- have gone better and this has highlight- nance was mentioned in the study text erence to senior management. In 2001, ed need to improve project manage- from 2001 onward, governance only there is reference to the CEO seeing the ment on a formal level.” Saying that began to appear as a theme in confer- organization as “becoming more proj- some “projects could have gone better” ence presentations in 2004, as did the ect focused” and in 2002 of “senior was an understatement, as the organi- Association for Project Management’s executives taking project management zation, like many others, had experi- guide to governance of project man- seriously.” Underlying this is an ongo- enced a number of significant failures. agement (Association for Project ing concern with retaining senior man- Throughout 2004, there was continued Management, 2004). agement support, a concern that is reference to assessment of competence shared and well documented else- to “get an organizational benchmark” IT/Software where (Thomas, Delisle, & Jugdev, and to providing a “development pro- Apart from reference to outsourcing of 2002). Therefore, any instance of gram for all individuals.” the IT division in 2000, there is no ref- expressed senior management support erence to IT/Software until 2002, when is reported as an achievement. Governance the “rollout” of a project to provide a Reference to governance begins in standard project management Organizational Learning 2001 and peaks in 2002. This timing is methodology to cover the “entire life Organizational learning comprises interesting as it coincides with a num- cycle of projects” across the organiza- both knowledge management and les- ber of regulatory initiatives aimed at tion is first mentioned. As this pro- sons learned. It is also associated with improving governance such as OECD gresses, there is mention in 2003 of project management community. (OECD, 2004) Principles of Corporate concern with “linking with ERP sys- There are several references to a “PM Governance, the Sarbanes-Oxley Act of tems… streamlining reporting, Forum,” set up in 2002, initially look- 2002, aimed at protecting investors by improving efficiency.” An intranet site ing at areas that were not being han- improving the accuracy and reliability is mentioned along with concern that dled well, such as project closings, and of corporate disclosures, and the Basel “PMs may not be aware of it, may find setting up groups to examine them. II New Accord, revising international it cumbersome, may be concerned that There is no indication to suggest that standards applicable to financial insti- it is all financially driven” because it these initiatives continued or had any tutions, released in January 2001 does not deal with such “project man- significant impact. (Basel Committee on Banking agement issues such as stakeholder The primary interest in knowledge Supervision, 2001). The focus on cor- management” and there is “confusion management appears focused on the porate governance, as part of a wider as to the process to follow.” Reflecting use of contractors and concerns about discourse, clearly has an influence on these concerns, there are subsequent minimizing the “external consultant the language and activities of people in references, on several occasions, to “re- spend and keep[ing] knowledge in- The Organization at this time. launch” of the intranet site. In 2004, house.” In 2004, they reported that “a Although governance must have been discussion of the intranet site and stan- small team of six people is looking at

82 AUGUST 2006 PROJECT MANAGEMENT JOURNAL setting up our own KM system.” As The Nature of Language and driven by the pressures of the with project management forums, The affect of business change in the moment; although, given the domi- there was no indication of the suc- “transformation” phase can be seen by nance of the concept in the organiza- cess or continuity of this initiative, looking beyond the organizational tional project management rhetoric, and, although there was one refer- project management themes to the there are no direct use of the term ence to recognition that projects nature of language used. The pace and “maturity” and/or any text that could might have been handled more effec- tone of the text change noticeably in be construed as referring to or imply- tively, there was very little evidence 2003. Prior to this, the language is rel- ing the concept. Of the organizational that organizational learning is a sus- atively placid. The terms “progressing project management capability-related tained interest. with,” “aiming for,” “revamping,” and terms (49 to 62 in Appendix A), the “looking at” are used on a number of only terms that do not appear at all, Text and Context occasions in the periods from 2000 either directly or by implication, are Business Change through 2002 in reporting on the “maturity” and “ethics/rules of con- This is clearly an important contextu- kinds of concerns and activities being duct.” Although there is no reference al issue, as there is reference to busi- undertaken. In 2003, the pace and to, or indication that there is, any ness change throughout the study nature of activity appears to increase. coherent “improvement program,” period. “Restructuring” occurs in There is a sense that the people there is a general sense of striving for 2000, 2001, and 2002. By 2003, the reporting are feeling a degree of pres- improvement, albeit with no clearly discourse has changed so that now, sure to perform which is expressed in stated baseline or goals. The lack of instead of “restructuring,” there is terms such as “fast track”—which is baseline and improvement goals is reference to “transformation.” The repeated on several occasions interesting, as The Organization did influence of this “restructuring” and throughout 2003 and 2004—and in use the Network tools to assess their “transformation” can be seen in vari- association with other terms such as organizational project management ous ways. First, it affects the voices the need to “fast track” to “achieve capability. But there is no evidence in that present the reports on current aggressive targets.” This sense of pres- the text to suggest that the feedback concerns and activities in terms of sure to perform is reinforced by from this was used to inform and organizational project management. phrases such as “accelerated delivery,” guide improvement. This is a point of As mentioned earlier, this shifted “delivering on promises,” “execution difference between the discourse with- from the IT division to the wider capability crucial,” and “charged with in The Organization and the discourse organization, initially driven by peo- increasing execution capability.” within other organizations involved in ple from internal audit although, on The phrases “looking at” and the Network, as analysis of the dis- one occasion, “four people from four “rolling out” appear regularly through- course in some other organizations different areas” attended, indicating out the entire study period and appear over the same period reflects a far “how diverse [The Organization] is to be characteristic of the ongoing dis- stronger and more coherent commit- in terms of project management.” course of organizational project man- ment to improvement of organization- When the EPO was formed, the peo- agement capability improvement, not al project management capability. The ple responsible for the EPO were the only within The Organization but reality of experience within The dominant voice, but in 2004, this among all the organizations involved Organization, however, demonstrates dominance was weakened and can with the Network. In a keyword analy- that despite expressed commitment to be seen in some of the conflicting sis of the text of reports from all organ- improvement, organizations do not messages being presented. izations in the network from 2000 to necessarily follow the step-by-step The presentation of business 2005, the phrase “looking at” was progress toward “maturity” of their change as “transformation,” as a con- highlighted as unusually frequent project management capability as pro- sistent theme in Discourse 2 from words in this body of text as compared moted in the literature and standards. 2003 onwards, reflects the wider dis- to a reference corpus. course of business, management, Surprisingly, given that the repre- Conclusion and organizational development. In sentatives of The Organization were, This paper has used the principles of a paper reviewing the major areas of by their membership in the Network, discourse analysis as a framework for focus in the literature base of the interested in improving organizational studying the extent to which practice organizational development field project management capability, there reflects the espoused theories of orga- (covering empirical findings, theory, were few references to improvement nizational project management capa- practice, applications, and interven- and no references to any coherent bility development. This has been tions) (Piotrowski & Armstrong, improvement program. There is no done by comparing two discourses, 2004) “organizational or transfor- sense, either implicit or explicit, of a one representing the espoused theories mational change” was identified as clear plan for improvement. The con- of project management practitioners the most popular area of study in the cerns, activities, and achievements that (Discourse 1) and the other represent- period from 1992 to 2003. are reported appear relatively ad hoc ing the reality of practice (Discourse 2).

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 83 The evolution of project management, in a relatively ad-hoc manner respond- accord: consultative document. [Web Page] as a field of practice brought into being ing to increasing pressure from senior Bank for International Settlements. through the conversations, writing, management to deliver desired bene- Retrieved November 18, 2006, from and collaborative activities of practi- fits. The path for development http://www.bis.org/publ/bcbsca03.pdf tioners, consultants, and academics appears opportunistic and highly sub- Betts, M., & Lansley, P. (1995). has been described as a journey from ject to changes in organizational International Journal of Project the conversations of senior practition- structure and priorities. Management: A review of the first ten ers and application of tools and tech- The influence of other discourses years. International Journal of Project niques on stand-alone projects to the is evident in the impact of demand for Management, 13(4), 207–217. conceptualization of project manage- higher standards of corporate gover- Chrissis, M. B., Konrad, M., & ment as an organizational capability. nance and in the effect of restructuring Shrum, S. (2003). CMMI®: Guidelines Themes and topics evident in the liter- and business change, also referred to as for process integration and product ature, standards, and guides have been “transformation,” reflecting a key pre- improvement. Boston: Addison Wesley identified as representative of the occupation of the organizational Professional. espoused theories of project manage- development field. Cooke-Davies, T.J., Schlichter, J., & ment practitioners (Discourse 1). The This in-depth study of the dis- Bredillet, C. (2001). Beyond the text of periodic reports on current con- course of organizational project man- PMBOK® guide: Creating a standard for cerns and project management agement capability development in organizational project management matu- improvement activities and achieve- one company over a four-year period rity. In Proceedings of the 31st Annual ments of one organization over a four- provides a very useful insight into the Project Management Institute 2001 year period have been analyzed as an reality of practice, and the extent to Seminars & Symposium, Nashville, TN, instance of practice (Discourse 2). which it reflects or may influence USA, CD-ROM. The underlying proposition is that espoused theories as embodied in Cooke-Davies, T. J. (2004). Project discourses are constructive and con- project management literature, stan- management maturity models. In P. W. stantly shaping, and being shaped by, dards, and guides. Although general- G. Morris & J. K. Pinto (Eds.), The their context and other discourses. ization is not possible on the basis of Wiley Guide to Managing Projects. Results of analysis demonstrate that the one case, additional studies along sim- Hoboken, NJ: John Wiley & Sons. discourse of practice (Discourse 2) ilar lines will be the subject of future Crawford, L., Pollack, J., & does reflect the majority of key themes research, leading to opportunities for England, D. (2006). Uncovering the specific to organizational project man- generalization across multiple cases. trends in project management: Journal agement capability represented in proj- emphases over the last 10 years. ect management literature, standards, Notes International Journal of Project and guides (Discourse 1, Appendix A). 1 The author was a member of the PMI Management, 24(2), 174-184. Analysis indicates that those taking an Standards Committee when this stan- Crawford, L. H. (2004a). Global organizational view of project manage- dards project was initiated. body of project management knowl- ment capability show little interest in edge and standards. In P. W. G. Morris topics and themes that have tradition- & J. K. Pinto (Eds.), The Wiley guide to ally been applied to individual or References managing projects. Hoboken, NJ: John stand-alone projects (e.g., time, cost, Alvesson, M., & Karreman, D. Wiley & Sons. and quality), and significantly more (2000). Varieties of discourse: On the Crawford, L. H. (2004b). interest in those topics and themes that study of organizations through dis- Supporting delivery capability: The reflect a wider organizational perspec- course analysis. Human Relations, many faces of project support in organ- tive. However, although the concept of 53(9), 1125–1149. isations. In Proceedings of 17th IPMA maturity is pervasive in Discourse 1, it Argyris, C. (1995). On organisa- World Congress on Project Management, does not appear either directly or indi- tional learning. Oxford: Blackwell June 2004. Budapest, Hungary: IPMA. rectly in practice (Discourse 2). Those Business. Dai, C. X., & Wells, W. G. (2004). engaged in the reality of organizational Argyris, C., & Schon, D.A. (1977). An exploration of project management project management capability devel- Theory in practice: Increasing professional office features and their relationship to opment are more concerned with capa- effectiveness. San Francisco: Jossey-Bass project performance. International bility and results than they are with the Publishers. Journal of Project Management, 22(7), concept of maturity. Reference to ethics Association for Project 523–532. and rules of conduct is similarly absent Management. (2004). Directing change. Dinsmore, P. C. (1996a). On the from the discourse of practice. In this A guide to governance of project manage- leading edge of management: particular text (The Organization), ment. High Wycombe, UK: Association Managing organizations by projects. there is no evidence of any coherent for Project Management. PM Network, 10(3), 9–11. plan for improvement. Instead, there Basel Committee on Banking Dinsmore, P. C. (1996b). Tom are a number of initiatives undertaken Supervision. (2001). The new Basel capital Peters is behind the times. PM

84 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Network, 10(9), 8–11. Pittsburgh, PA: Software Engineering industry benchmark. Project Management Dinsmore, P. C. (1996c). Toward Institute, Carnegie Mellon University. Journal, 34(1), 4–11. corporate project management: Kerzner, H. (1979). Project manage- Piotrowski, C., & Armstrong, T. R. Beefing up the bottom line with ment: A systems approach to planning, sched- (2004). The research literature in organi- MOBP. PM Network, 10(6), 10–13. uling and controlling (1st ed.). New York: zation development: Recent trends and Dinsmore, P. C. (1996d). What if the Van Nostrand Reinhold. current directions. Organization CEOs find out?: The strategic positioning Kloppenborg, T. J., & Opfer, W. A. Development Journal, 22(2), 48. of project management. PM Network, (2000). Forty years of project manage- Project Management Institute. 10(12), 8–11. ment research: Trends, interpretations, (2003). Organizational Project Management Dinsmore, P. C. (1999). Winning with and predictions. In Project Management Maturity Model: Knowledge Foundation. enterprise project management. New York: Research at the Turn of the Millennium: Newtown Square, PA: Project AMACOM. Proceedings of PMI Research Conference (pp. Management Institute, Inc. Gareis, R. (1992a). The handbook of 41–59), 21 - 24 June 2000, Paris, France. Shenhar, A. J. (1998). From theory to management by projects. Vienna: Manz. Koskela, L., & Howell, G. (2002). The practice: Toward a typology of project Gareis, R. (1992b). Management of underlying theory of project management management styles. IEEE Transactions on networks of projects. 1992 AACE is obsolete. In: D. P. Slevin, D.I. Cleland, & Engineering Management, 45(1), 33–48. Transactions, Volume 2, American J. K. Pinto (Eds.), Frontiers of Project Stretton, A. (1994a). A short history Association of Cost Engineers, Morgantown, Management Research and Application: of project management: Part one: the West Virginia, USA, J.1.1 – J.1.6. Proceedings of PMI Research Conference (pp. 1950s and 60s. The Australian Project Graham, R. J., & Englund, R. L. 293–301), July 2002, Seattle, WA, USA. Manager, 14(1), 36–37. (1997). Creating an environment for success- Morris, P.W.G. (1994). The management Themistocleous, G., & Wearne, S. H. ful projects: The quest to manage project man- of projects (1st ed.). London: Thomas Telford. (2000). Project management topic cover- agement. San Francisco: Jossey-Bass Morris, P.W.G. (2000). Researching age in journals. International Journal of Publishers. the unanswered questions of project man- Project Management, 18(1), 7–11. Hardy, C. (2001). Researching orga- agement. In D. P. Slevin, D.I. Cleland, & J. Thomas, J., Delisle, C., & Jugdev, K. nizational discourse. International Studies K. Pinto (Eds.), Project Management (2002). Selling project management to senior of Management & Organization, 31(3), Research at the Turn of the Millennium: executives: Framing the moves that matter. 25–47 Proceedings of PMI Research Conference (pp. Newtown Square, PA: Project Hildreth, P. (2000). Communities of 87–101), 21 - 24 June 2000, Paris, France. Management Institute. practice. Retrieved April 7, 2000, from Morris, P. W. G., Patel, M. B., & Urli, B., & Urli, D. (2000). Project http://www-users.cs.york.ac.uk/~pmh/ Wearne, S.H. (2000). Research into revis- management in North America, stability work.html#Communities ing the APM project management body of of the concepts. Project Management Hobbs, J. B., & Aubry, M. (2005). A knowledge. International Journal of Project Journal, 31(3), 33–43. realistic portrait of PMOs: The results of Management, 18(3), 155–164. van Dijk, T. A. (1997). The study of an empirical investigation. In Proceedings OECD. (2004). OECD Principles of discourse. In T. A. van Dijk (Ed.) Discourse of PMI Global Congress North America, Corporate Governance. (Rev. ed.). Paris: as structure and process—Discourse studies: A Toronto, Canada. Newtown Square, PA: OECD. multidisciplinary introduction. (pp. 1–34). Project Management Institute. Parker, I. (1992). Discourse dynamics. London: SAGE Publications Ltd. Hodgson, D. E. (2004). Project work: London: Routledge. Wenger, E. C., & Snyder, W. M. The legacy of bureaucratic control in the Paulk, M. C., Weber, C. V., Curtis, B., (2000). Communities of practice: The post-bureaucratic organization. & Chrissis, M. B. (1995). The capability organizational frontier. Harvard Business Organization, 11(1), 81–100. maturity model: Guidelines for improving the Review, 78(1), 139–146. Humphrey, W. S., & Sweet, W. L. software process. Boston: Addison-Wesley Zobel, A. M., & Wearne, S. H. (2000). (1987). A method for assessing the software Longman Publishing Co., Inc. Project management topic coverage in engineering capability of contractors: Pennypacker, J.S., & Grant, K.P. recent conferences. Project Management CMU/SEI-87-TR-23, ESD/TR-87-186. (2003). Project management maturity: An Journal, 31(2), 32–37.

LYNN CRAWFORD is a professor of project management at ESC Lille, France Director, Project Management Research Group, University of Technology, Sydney Managing Director, Human Systems Pty Ltd. She is involved in project management education, practice, and research. Through human systems, she works with leading corporations that are developing organizational project management competence by sharing and developing knowledge and best practices as members of a global system of project management knowledge networks. She is currently involved in two PMI-funded research projects—Exploring the Role of the Executive Sponsor and The Value of Project Management. Results of a completed study have been published in 2005, in a book titled Project Categorization Systems: Aligning Capability with Strategy for Better Results. Lynn has directed major research projects on project management competence and soft systems for soft projects.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 85 Appendix A: Codes Used in Text Analysis

48 Topics/Themes from PM Literature, Standards and Selected Organizational PM Context Guides (Crawford, 2004) Capability-Related Terms

1. Benefits management 25. Project closeout/ 49. Benchmarking 63. Business change/ 2. Business case finalization 50. Improvement programs restructuring 3. Change control 26. Project context/ 51. Ethics/rules of conduct 64. Contractors environment 4. Configuration 52. Culture 65. IT/software 27. Project evaluation review management 53. Governance 66. Outsourcing 28. Project initiation/start-up 5. Conflict management 54. Management by projects 6. Cost management 29. Project life cycle/ project phases 55. Maturity 7. Design management 30. Project monitoring 56. Community 8. Document management control 57. Competence career 9. Estimating 31. Project organization development 10. Financial management 32. Project planning 58. Policies, processes, methodology tools 11. Goals, objectives and 33. Quality management strategies 59. PMO/support office 34. Regulations 12. Information/ 60. Portfolio management 35. Reporting communication 61. Sponsorship* management 36. Requirements 62. Top management support 13. Integration management management 14. Leadership 37. Resource management 15. Legal issues 38. Risk management 16. Marketing 39. Safety, health and environment 17. Negotiation 40. Stakeholder/relationship 18. Organizational learning management (inc. Lessons) 41. Strategic alignment 19. Performance measurement (inc. EVM) 42. Success 20. Personnel/human 43. Team building/ resource management development/teamwork 21. Problem solving 44. Technology management 22. Procurement 45. Testing, commissioning handover/acceptance 23. Program/programme management 46. Time management/ scheduling/phasing 24. Project appraisal 47. Value management 48. Work content and scope management

Note: Items shown in italics above are more directly related to organizational project management capability than to the management of individual projects.

86 AUGUST 2006 PROJECT MANAGEMENT JOURNAL PROJECT-BASED MANAGEMENT AS AN ORGANIZATIONAL INNOVATION: DRIVERS, CHANGES, AND BENEFITS OF ADOPTING PROJECT-BASED MANAGEMENT

MIIA MARTINSUO, Helsinki University of Technology NICOLE HENSMAN, University of Sydney KARLOS ARTTO, Helsinki University of Technology JAAKKO KUJALA, Helsinki University of Technology ALI JAAFARI, Asia-Pacific International College

Project-Based Management as an Organizational Innovation ABSTRACT ne track of innovation management literature examines innovations that change the ways in which the organization operates. Organizational inno- This paper examines project-based man- O vation can be considered an idea or behavior new to the adopting organi- agement as an organizational innovation. Institutional theory and innovation diffu- zation (Damanpour & Evan, 1984). Organizational innovation may encompass sion literature suggest that the drivers for new products or services, new process technologies, new organizational structures adopting an organizational innovation or administrative systems, or new plans or programs pertaining to organizational may differ across organizations, and that members (Alänge, Jacobsson, & Jarnehammer, 1998; Damanpour, 1996; the drivers may be linked with the timing Damanpour & Evan, 1984). The idea may be internally generated or borrowed of the innovation. A survey questionnaire was used for data collection, and the sam- from other organizations (Damanpour & Evan). ple consisted of 111 companies represent- Project-based management can be considered an organizational innovation ing a variety of industries. The results of that may influence both the technical and social system of the organization this study identified external pressure and through new structures, methods, technical systems, and behavioral patterns. internal complexity as drivers for introduc- Project-based management has at least four special features, as compared to other ing project-based management. The degree of process change, depth of proj- forms of management. ect-based management adoption, and 1. Project-based management is directed toward organizing activities to local success of project-based manage- achieve goals of scope, cost, and time (PMI, 2004; Turner, 1999) and, ment introduction as changes caused by increasingly, toward broader customer and business goals (Shenhar, Dvir, adopting project-based management are Levy, & Maltz, 2001). In earlier research, management by objectives has examined. The study also reveals benefits from introducing project-based manage- been considered as an organizational innovation, as well as goal-oriented ment in the form of improvement in proj- programs (Fennell, 1984). ect culture, and efficiency improvement. 2. Project-based management induces a temporary organizational structure as part of or replacing the old organizational structure (Packendorff, 1995; Keywords: innovation management; PMI, 2004). Earlier, M-form or matrix organizational structures (Burns & organizational innovation; project-based management Wholey, 1993; Mahajan, Sharma, & Bettis, 1988; Teece, 1980), and flow manufacturing in multiple plants (Maritan & Brush, 2003) have been ©2006 by the Project Management Institute examined as organizational innovations. Vol. 36, No. 3, 87-97, ISSN 8756-9728/03 3. Project-based management can include both standardized and organization-specific tools and good practices (Milosevic & Patanakul, 2005; PMI, 2004; White & Fortune, 2002). As a comparison, studies on total quality management (TQM) and ISO 9000 have earlier been considered as organizational innovations (Guler, Guillén, & Macpherson, 2002; Westphal, Gulati, & Shortell, 1996; 1997). Also data processing and IT solutions have been studied (Kimberly & Evanisko, 1981).

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 87 4. Project-based management pro- mentation of new ideas or behaviors in their rationale (motives or drivers). motes distributed and project- or across organizations. Earlier research Innovation diffusion literature sug- specific responsibilities in the on institutionalization suggests that gests that firms adopt new organiza- organization (PMI, 2004; Turner, those companies that adopt an organi- tional innovations to maintain and 1999). Each project has a dedicat- zational innovation early have more enhance their performance ed project manager and project freedom to modify practices to increase (Damanpour, 1987)—e.g., ensure cost organization that dissolves as the organizational efficiency. Later effective production of high quality project ends. New management adopters, in turn, have normative pres- products and services. Institutional the- system has earlier been consid- sures to comply with the practices ory provides a complementary ration- ered as organizational innovation developed by the early adopters ale: new innovations are adopted to in somewhat different settings, (Westphal et al., 1997). This mecha- ensure social fitness and legitimacy i.e., administrative process or staff nism of institutional isomorphism (Meyer & Rowan, 1991). Organizations development program in libraries may force organizations to take into that adopt organizational innovations (Damanpour, 1987; Damanpour use practices that do not fit with the increase their legitimacy and survival & Evan, 1984; Damanpour, organization (e.g., DiMaggio & Powell, prospects regardless of the efficiency of Szabat, & Evan, 1989). 1983). As a result, the way in which the adopted practices (Meyer & the organizational innovation is intro- Rowan). Institutional theory differenti- A specific feature of organization- duced and, thereby, adopted may ates between the early and late adopters al innovations is that the “product” is influence the degree to which it suc- of innovations and have identified not as clear as in other types of inno- ceeds in bringing about sustainable somewhat different drivers for these vations, and the incentives for devel- competitive advantage. groups. For example, research on the oping them are not immediately Project management research has adoption of TQM in the healthcare sec- apparent (Alänge et al., 1998). Yet, covered development of project-based tor has reported that earlier adopters organizational innovations have been management through different maturi- implemented TQM mainly to increase considered particularly important and ty models, competency models, excel- efficiency and effectiveness of their interesting for the survival and success lence models, and scorecards (e.g., work processes (Westphal et al., 1997). of the firm. For example, Powell Andersen & Jenssen, 2003; Cooke- They modified the innovation for their (1995) has examined TQM as an orga- Davies & Arzymanow, 2003; Cormican own needs and integrated new prac- nizational innovation and proposed & O’Sullivan, 2004; Ibbs & Kwak, 2000; tices into the working processes. Later that it contributes to a sustained com- Jugdev & Thomas, 2002; Kwak & Ibbs, adopters rather focused on the symbol- petitive advantage. Due to the coexis- 2002; Westerveld, 2003). Many of such ic benefits of such an innovation. Our tence of the previously mentioned studies examine the maturity or com- first research question is: What are the features, project-based management petence areas relevant to successful main drivers for introducing project-based can be considered a highly interesting project-based management, differences management? organizational innovation that may across firms or industries, and the steps Second, the actual changes caused face difficulties when being adopted through which companies develop by the adoption process are relevant in and developed. their project-based management. Such determining whether adoption has studies often assume that project-based taken place or not. Damanpour and Adopting Organizational Innovations management is already in use and that Evan (1984; also Damanpour, 1987) One inherent feature of organizational companies differ in their maturity of pondered whether an innovation has innovations is that their imitation, i.e., project-based management. The origi- been adopted upon its decision, start spread in and across organizations, is nal introduction or adoption of project- of implementation, or only after suc- difficult if not impossible due to based management has received little cessful implementation. They conclud- organization-specific implementation attention. Therefore, utilizing innova- ed that the idea can be considered conditions, and the local interpretation diffusion and institutional theory adopted (well or poorly) only when tions necessary (e.g., Mahajan et al., to better understand the early phases the idea is actually being used. 1988; Teece, 1980). However, compa- and diffusion of project based manage- Abrahamson (1991) has drawn atten- nies do attempt to imitate organiza- ment in and across firms could con- tion to the fact that some innovations tional innovations because they seek tribute to project management are actually rejected, and that organiza- the same benefits as many other firms research. Both innovation and institutional processes sometimes prompt and because there are no obvious pro- tional theory literature encourage the adoption of inefficient innova- tective barriers such as patenting examining three topics: rationale for tions, besides the efficient ones. Both opportunity for the use of these inno- project-based management, and the Damanpour’s and Abrahamson’s stud- vations (Teece, 1980). changes, and benefits stemming from ies suggest that the actual changes Adoption of organizational inno- adopting project-based management. accomplished through adopting an vations is a process that includes the First, companies may differ in the organizational innovation are related generation, development, and imple- adoption of an innovation in terms of to the drivers and adoption conditions

88 AUGUST 2006 PROJECT MANAGEMENT JOURNAL (also Alänge et al., 1998; Kimberly & This paper covers only those survey Questionnaire Items Included in Evanisko, 1981). Our second research topics that focus on the introduction the Analysis question is: What kind of changes has of project-based management. For the purposes of this study, we used the introduction of project-based manage- 23 questionnaire items to examine the ment caused in practices and processes in Survey Sample drivers, changes and benefits related to the organization, and are these changes The survey was carried out across introducing project-based management. associated with the drivers? Australian companies representing a Benefits of Introducing Project- Third, the benefits or outcomes of variety of industry sectors. Originally, Based Management. The survey asked: adopting an innovation can be consid- the questionnaire was mailed to 4,800 What benefits has project-based man- ered relevant. Abrahamson (1991) companies based on Australian agement brought to your company? noted the proinnovation bias in inno- Business Review listing of top firms in Eight items were included: greater vation diffusion research: the domi- the country. Of these, 111 companies entrepreneurship, more client satisfac- nant assumption is that innovation is responded to the survey, with a total tion, more effective communication, always brought to completion and response rate of 2.3%. A number of more knowledge management and would benefit adopters. In reality, people in the original target popula- know-how transfer, improved project good innovations may be rejected and tion reported lack of project-based control, better multiproject coordina- bad ones adopted. Fads and fashions management in their organization and, tion, greater project transparency, and may promote even quite unbeneficial therefore, non-response. The low better project performance. A scale of 1 innovations in uncertain environ- response rate may also have resulted (completely disagree) to 5 (totally agree) ments, and encourage imitation across from the rather heavy questionnaire was used. organizations. Earlier research indi- form, and another survey on the same Changes Through Introducing cates that some aspects of the innova- population being launched at the same Project-Based Management. We tion diffusion process, e.g., time. The sample characteristics indi- examined changes in three areas: standardization of the innovation, or cate a skewedness toward rather experi- degree of process change, depth of the use of an external, trustworthy enced project personnel, which may project-based management adoption, institution of expertise, may impact influence the results and need to be and local success of project-based both the adoption of the innovation, considered as a limitation of the study. management introduction. Degree of and the associated benefits (Alänge et A majority of the responses were process change examined how much al., 1998; Fennell, 1984; Westphal et received from public sector and service the work processes changed as a result al., 1997). Our third research question organizations, with a minority of of introducing project-based manage- is: What are the perceived benefits from responses representing the more tradi- ment for your area, for your depart- introducing project-based management, tional project businesses such as capi- ment, and for you personally. A scale and how are these benefits associated with tal industry, manufacturing, and IT of 1 (hardly at all) to 5 (a great deal) the drivers and changes? and telecommunications. Small to was used. Three questions in depth of The purpose of this research is to medium-sized companies dominate in project-based management adoption examine project-based management as the sample. A majority of responses examined the presence of project-based an organizational innovation. More come from firms where project man- management: project-based manage- specifically, we study the drivers, agement has been officially introduced ment culture is widely present at all lev- changes and benefits of introducing throughout the firm. The time of intro- els of the hierarchy, project-based project-based management, and link- ducing project-based management management is used sporadically in the ages between them. varies strongly. Background informa- company (scale inverted for further tion on the companies participating in analyses), and the project and line Research Method the survey is presented in Table 1. organizations work well together in the A questionnaire survey was used to Of the respondents, 80% are company. A scale of 1 (completely dis- examine the introduction, current state male, and 27% are members of a agree) to 5 (totally agree) was used. Local and future prospects in project-based project management association. success of project-based management management. The questionnaire was The respondents represent different introduction was measured with two originally developed in Germany age groups, have dominantly an items that asked: How successful was (Volkswagen Coaching, 2002) and economics or business education, the introduction of project-based man- later adopted by other countries. The and a majority represent project agement in your area? How successful Australian version is a modification of management or finance tasks. The was the introduction of project-based the original survey, developed further respondents are very experienced management in your department? The on the basis of expert interviews and in project work, i.e., over 53% have items used a scale of 1 (unsuccessful) to literature review (the background and more than 10 years of experience in 5 (very successful). methodology of the entire research is projects. More information on the Drivers for Introducing Project- explained more thoroughly in individual respondents is present- Based Management. The question- Hensman, Valenta & Jaafari, 2004). ed in Table 2. naire asked for the main reasons for

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 89 Industry % Number of Employees % 1 = Public sector 30.6 1 = Below 500 51.4 2 = Services 28.8 2 = 501-1000 16.2 3 = Manufacturing 15.3 3 = 1001-5000 21.6 4 = IT and telecommunications 15.3 4 = Over 5001 10.8 5 = Capital industries (construction, energy etc.) 9.9

Was PM Officially Introduced? % Years from Introducing PM % 1 = No, it was not officially introduced at department 29.7 1 = Below 1 6.3 or company level 2 = Yes, it was officially introduced either at 27.0 2 = 1-3 years 35.1 department level or throughout company 3 = Yes, it was officially introduced both at 38.7 3 = 4-10 years 28.8 department level and at company level n.a. 4.5 4 = Over 10 years 23.4 n.a. 6.3

Table 1: Companies in the survey sample (N=111)

Age Group % Training Background % Area of Work % Years in Project Work % Below 34 21.6 Economics, business 51.4 Project management 36.0 1 = Below 1 0.9 35-44 33.3 Engineering 15.3 Finance, accounting 20.7 2 = 1-3 8.1 45-54 38.7 IT 12.6 Internal consultancy, 12.6 3 = 3-10 37.8 staff Over 55 6.3 Science 9.0 Information technology 10.8 4 = Over 10 53.2 Other 11.7 Strategy, planning, 6.3 development Other 13.5

Table 2: Individual respondents’ background (N=111)

introducing project-based manage- years of involvement with project work the factors explain 76% of the variance ment. Seven items were used: increasing (ordinal scale, as in Table 2). in the model. For the benefit items, we project complexity, increasing number used oblique (direct oblimin) rotation of projects, time pressure for projects, Preliminary Analysis and due to expected item intercorrelations. image of modernity, client demands, Descriptive Statistics A two factor model was supported for internationalization and globalization, To explore and identify the variable “benefits” and explains 69% of the and market or competitive pressure. structure, we conducted principal com- variance. We named the benefit vari- These items were measured on a scale of ponents analysis of the items and tried ables as improvement of project culture 1 (unimportant) to 5 (very important). out different models. For the drivers and efficiency improvement. Two items Control Variables. We used four and changes we used orthogonal (vari- have fairly high component loadings control variables at company level, all max) rotation. A two-factor model is outside of the proposed variable struc- introduced in Table 1: industry, num- suggested for “drivers”: internal com- ture, as shown in Appendix 1. ber of employees, official introduction plexity and external pressure, and the fac- However, we chose to include them as of project-based management, and tors account for 57% of the variance in part of the principal component factor. years from introducing project-based the model. For “changes,” a three-fac- We developed variables based on management. Additionally, we con- tor model was used as indicated by the the principal components for further trolled two individual level variables: question setting: degree of process analysis. Scores for each variable membership of an association for proj- change, depth of project-based manage- were calculated as average of the ect management (dummy variable, ment adoption, and local success of proj- included items. To estimate the relia- 1=member, 0=non-member), and ect-based management introduction, and bility of the variables, Cronbach’s

90 AUGUST 2006 PROJECT MANAGEMENT JOURNAL alpha coefficients were calculated. The based management, timing of proj- linear relationship between independ- scores for benefit and change variables ect-based management introduction ent and dependent variables. We tried are high (0.73 - 0.89), but slightly below and project management association different models and decided to use a the acceptable level of 0.7 for driver vari- membership have a few correlations four-step regression approach for both ables (0.66 and 0.68). The content and with the other variables. For instance, the dependent variables. First, we reliability coefficients for the variables traditional project industries have a entered the control variables (model 1), are presented in Appendix 1. longer history with project-based then we added the drivers (model 2), Means, standard deviations, and management than does public sector third we added a degree of process correlation coefficients among the and services. Traditional project change and depth of project-based variables are presented in Table 3. industries also report external pres- management adoption (model 3), and Internal complexity dominates as a sure more often as a driver for intro- finally we added the local success in driver for introducing project-based ducing project-based management, introducing project-based manage- management, as compared to external and higher depth of project-based ment (model 4). Model 1 (i.e., control pressure. Of the change variables, management adoption. The driver, variables alone) did not prove suffi- local success of project-based man- change, and benefit variables have a cient for explaining variance in either agement introduction receives slightly number of significant correlations of the dependent variables. higher scores, as compared to degree with each other. of process change or depth of project- Improvement of Project Culture based management adoption. Of the Drivers, Changes, and Benefits of Models 2–4 are suitable for explaining benefit variables, the score of efficien- Adopting Project-Based Management variance in “improvement of project cy improvement is somewhat higher To better understand the links between culture,” but especially model 2 has a than improvement of project culture. drivers, changes, and benefits, we con- low explanatory value. In model 2, the Of the control variables, indus- ducted linear regression analysis on control variables and drivers together try, official introduction of project- the variables. The scatterplots revealed explain only 18% of variance in N N mean s.d. Industry Number of employees Official introduction of PM from introducing PM Years Are you member of an association for PM in project work Years Internal complexity External pressure Degree of process change Depth of PM adoption Local success of PM introduction Improvement of project culture

Industry 111 2.45 1.33 Number of employees 111 1.92 1.08 -0.14 Official introduction of PM 106 1.09 0.85 -0.09 -0.05 Years from introducing PM 104 2.74 0.91 0.28** 0.12 -0.15 Are you member of an 111 0.27 0.45 0.02 -0.11 0.09 -0.02 association for PM Years in project work 111 3.43 0.68 0.19* 0.01 -0.12 0.45*** 0.09 Internal complexity 111 4.29 0.62 -0.01 0.13 0.06 0.08 -0.02 0.16 External pressure 111 2.67 0.90 0.28** -0.06 0.06 0.21* 0.20* 0.12 0.33*** Degree of process change 111 3.32 0.89 -0.13 0.13 0.05 -0.22* 0.05 -0.13 0.07 0.13 Depth of PM adoption 111 2.97 0.99 0.39*** -0.09 0.21* 0.30** 0.27** 0.27** 0.10 0.44*** -0.06 Local success of PM 108 3.63 0.97 0.05 0.07 0.27** 0.00 0.15 0.17 0.26** 0.23* 0.33*** 0.41*** Introduction Improvement of project 111 3.25 0.75 0.12 -0.02 0.07 0.10 0.22* 0.08 0.17 0.47*** 0.19* 0.44*** 0.39*** culture Efficiency improvement 111 3.78 0.76 0.05 -0.02 0.13 0.06 0.18 0.05 0.22* 0.30** 0.25** 0.38*** 0.58*** 0.63***

***p<0.001, **p<0.01, *p<0.05 Table 3: Descriptive statistics and correlation coefficients (PM = project-based management)

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 91 Industry 0.13 1.28 0.03 0.26 -0.05 -0.50 -0.03 -0.29 Number of employees 0.05 0.45 0.05 0.51 0.05 0.58 0.03 0.36 Official introduction of PM 0.08 0.80 0.04 0.42 -0.06 -0.67 -0.12 -1.34 Years from introducing PM 0.05 0.41 -0.03 -0.26 -0.08 -0.74 -0.05 0.54 Are you member of an 0.23 2.25* 0.14 1.50 0.05 0.61 0.06 0.68 association for PM Years in project work 0.05 0.41 0.03 0.31 -0.01 -0.10 -0.06 -0.64 Internal complexity -0.01 -0.11 -0.01 -0.12 -0.05 -0.54 External pressure 0.44 4.06*** 0.25 2.41* 0.26 2.57* Degree of process change 0.14 1.62 0.05 0.50 Depth of PM adoption 0.47 4.41*** 0.37 3.35** Local success of PM 0.28 2.68** Introduction R2 0.09 0.25 0.4 0.45 Adjusted R2 0.03 0.18 0.34 0.38 F 1.53 3.73** 5.93*** 6.43***

Standardized Beta coefficients are shown ***p<0.001, **p<0.01, *p<0.05

Table 4: Regression analysis, improvement of project culture as dependent variable

improvement of project culture. and the dependent variable remains ciency improvement. In model 3, External pressure appears as a strong almost significant. degree of process change and depth of and significant contributing variable. In model 4, altogether 38% of project-based management adoption The more external pressure is experi- variance in the dependent variable is both appear as significant and fairly enced as a driver for introducing proj- explained. Local success of project- strong variables, and the model ect-based management, the more based management introduction explains altogether 24% variance in improvement is seen in project culture. appears as a significant variable, slight- efficiency improvement. That is, higher Internal complexity as a driver for ly mediating the impact of depth of degrees of process change and wide, introducing project-based manage- project-based management adoption consistent, and thorough use of project ment, however, does not explain vari- and external pressure. If the introduc- management are reflected in higher ance in the benefit “improvement of tion of project management is per- perceived efficiency improvements. project culture.” Table 4 reports the ceived as successful locally, also Table 5 shows the regression analysis regression analysis for improvement of improvements in project culture are results for efficiency improvement. project culture. perceived high. Model 4 explains 39% variance in Model 3 explains 34% variance in Control variables do not appear efficiency improvement. Again, local improvement of project culture and as significant, besides project man- success of project-based management shows that depth of project-based agement association membership in introduction is significant. It clearly management adoption adds explana- model 1. This effect is removed in the mediates the relationship between the tory power and is a significant variable. other models, indicating that the other change variables and efficiency This means that wide, consistent, and relationship between association improvement: the impact of degree of thorough use of project management membership and improvement of process change on the dependent vari- is reflected in perceived improvements project culture is mediated by drivers able is largely explained through local in project culture in terms of entrepre- and changes. success of project-based management neurship, knowledge transfer, client introduction, and also depth of proj- satisfaction, and communication. Efficiency Improvement ect-based management adoption has a Also, depth of project-based manage- Only models 3 and 4 are suitable for clearly lower score than in model 3. ment adoption and degree of process explaining variance in efficiency change seem to mediate the relation- improvement, and model 3 has still a Discussion ship between external pressure and the fairly low explanatory value. Control The results of this study have identified dependent variable, but also a direct variables and drivers alone or together external pressure and internal com- relationship between external pressure do not explain much variance in effi- plexity as drivers for introducing proj-

92 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Model 1 Model 2 Model 3 Model 4 t t t t Industry 0.05 0.50 0.00 0.04 -0.06 -0.61 -0.02 -0.28 Number of employees 0.06 0.56 0.04 0.37 0.02 0.25 -0.01 -0.13 Official introduction of PM 0.17 1.64 0.14 1.36 0.05 0.57 -0.05 -0.55 Years from introducing PM 0.03 0.24 0.00 -0.04 -0.01 -0.08 0.03 0.31 Are you member of an 0.17 1.64 0.13 1.25 0.05 0.52 0.06 0.66 association for PM Years in project work 0.10 0.81 0.06 0.53 0.05 0.44 -0.05 -0.46 Internal complexity 0.12 1.06 0.11 1.07 0.04 0.45 External pressure 0.21 1.79 0.01 0.06 0.02 0.22 Degree of process change 0.29 2.99** 0.11 1.23 Depth of PM adoption 0.42 3.65*** 0.24 2.18* Local success of PM 0.48 4.74*** Introduction R2 0.07 0.14 0.32 0.46 Adjusted R2 0.01 0.06 0.24 0.39 F 1.22 1.82 4.11*** 6.69***

Standardized Beta coefficients are shown ***p<0.001, **p<0.01, *p<0.05

Table 5: Regression analysis, efficiency improvement as dependent variable

ect-based management (research ques- linear relation with some change and management adoption, and local suc- tion 1). The choice of introducing proj- benefit variables. External pressure in cess of project-based management ect-based management is dominantly these results may represent access to introduction as changes caused by motivated by increased degrees of inter- knowledge or collaboration in a wider adopting project-based management nal complexity. Although the respon- network (Alänge et al., 1997; Fennell & (research question 2). Our results dents seem to have rationalized the Alexander, 1987; Westphal et al., emphasize that achieving benefits introduction of project-based manage- 1997), strategic business and customer from project-based management ment with the intent to control com- benefit expectations (Shenhar et al., requires both a wide, consistent, and plexity, this is not related to the changes 2001), and possibly also isomorphic thorough use of project management achieved by project-based manage- pressures from institutions in the same throughout the firm, and local success ment, or its benefits. As internal com- area or industry (Abrahamson 1991; in introducing project-based manage- plexity and external pressure correlate, it DiMaggio & Powell, 1983; Meyer & ment. This is in line with an earlier is possible that internal complexity is Rowan, 1991) that can be connected proposition that a great number of connected to changes and benefits, but with the adoption of the organization- units should support the use of the with a time lag. Largely in line with ear- al innovation. Innovation diffusion lit- organizational innovation, and that a lier literature, our sample may represent erature has posited that such an localized search process is needed for (dominantly) early adopters who have “efficient choice” perspective prompts the organizational innovation to suc- proactively sought to adopt project- companies to adopt innovations that ceed (Alänge et al., 1998). The results based management to solve their effi- close their identified performance gaps show that depth of project-based ciency and effectiveness concerns. The (Abrahamson, 1991). Our results did management adoption and local suc- sample characteristics may, therefore, not find direct evidence on the differ- cess of project-based management explain the missing connection from ences between early adopters and lag- introduction correlate with external internal complexity to changes and gards suggested by innovation pressure, and mediate the relationship benefits. Furthermore, for instance diffusion and institutional theory; between external pressure and the Damanpour (1987) reported that orga- however, this may indicate more com- benefit variables. nizational complexity can explain vari- plex, path-dependent relationships Degree of process change has only ance in technical innovations, but not between the timing of project-based an intermediary role toward achieving so much administrative (organization- management introduction, drivers, the benefits of project-based manage- al) innovations. and other variables. ment through local success. While the In turn, external pressure as a driv- We examined the degree of degree of process change may reflect the er had a lower average score but was in process change, depth of project-based degree of adoption of project-based

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 93 ment as shown in Figure 2. Although the drivers do not appear to have a significant role, degree of process change Benefit: Depth of PM has an indirect link to efficiency Improvement of Adoption improvement through local success of Project Culture External introducing project-based manage- Pressure ment. This finding indicates that process change as such is not self-evi- Local dently beneficial but, rather, must be Success of approved and adjusted at the local set- Introducing ting. These findings suggest that the PM linkages from the studied drivers to efficiency improvement are mediated Internal by some other variables, or that effi- Complexity Degree of ciency improvements are originally Process driven by some other forces than those Change covered in our study. Strategic choices, top management support, pressures from outside institutions, project man- Figure 1: Factors contributing to improvement of project culture as a benefit from introducing agement standardization, or the prac- project-based management tices used while introducing project-based management are exam- management, our results highlight the based management adoption are ples of possible relevant factors. importance of a subjective estimate mediated. These findings emphasize Earlier studies suggest and report of those changes, to be perceived as the necessity to locally adjust and some individual and organizational back- beneficial. Additionally, the results modify (local success) the thorough, ground variables relevant to the adoption may suggest that other kinds of company-wide solution (depth of of organizational innovations For exam- changes should be studied besides project-based management adoption) ple, organizational complexity and size process change (e.g., attitude and to reap the practical benefits of project- have been considered among significant behavioral changes may be equally based management. background variables (e.g., Damanpour, important for the adoption of project A significant degree of variance in 1996; Kimberly & Evanisko, 1981). Our based management). Addition of efficiency improvement is explained results did not directly confirm such find- such variables could have improved through depth of project-based man- ings. This may be explained through how the explanatory power of our regres- agement adoption, and local success of our background question was set: small, sion models. The relationship introducing project-based manage- medium, and almost large firms were between internal complexity, external pressure and degree of process change was not revealed with our analysis (i.e., the relationship could be Depth of PM nonlinear or more complex). Adoption The study revealed benefits from External introducing project-based manage- Pressure ment in the form of improvement in project culture, and efficiency improvement (research question 3). Even if the Local Benefit: items are strongly intercorrelated, their Success of Efficiency relation with drivers and changes of Introducing Improvement introducing project-based manage- PM ment are somewhat different. A signif- Internal icant degree of variance in Complexity improvement of project culture is Degree of explained by external pressure, depth Process of project-based management adop- Change tion, and local success of project-based management introduction as depicted in Figure 1. Part of the impact of exter- Figure 2: Factors contributing to efficiency improvement as a benefit from introducing nal pressure and depth of project- project-based management

94 AUGUST 2006 PROJECT MANAGEMENT JOURNAL included in a single response category Limitations Andersen, E. S., & Jenssen, S. A. (below 500 persons) and may have The generalizability of the results of this (2003). Project maturity in organizations. International Journal of Project blocked out the most influential differ- study is weakened by some limitations Management, 21, 457–461. ences, and we did not use other complexi- regarding the sampling, survey design, Burns, L. R., & Wholey, D. R. ty variables. Also, our analysis setup did and analysis setup. We have reported the (1993). Adoption and abandonment not fully uncover the relationship between sampling procedure, low response rate, of matrix management programs: control variables and changes, which and possible skewedness in the sample as Effects of organizational characteristics could be examined more in future studies. compared to the whole population. and interorganizational networks. Despite these limitations, we succeeded Academy of Management Journal, 36(1), 106–138. Ideas for Further Research in having very diverse firms in different Cooke-Davies, T. J., & Arzymanow, To confirm the findings, more elaborate industries as part of the sample. The sam- A. (2003). The maturity of project models should be developed on project- ple size of more than 100 is already management in different industries: based management as an organizational appropriate for statistical testing. Even if An investigation into variations innovation. Besides confirmatory analyses the data cannot fully cover the current between project management models. on our findings, additional research ques- state of project-based management in International Journal of Project Management, 21, 471–478. tions have been identified. For instance, Australian firms, the findings with these Cormican, K., & O’Sullivan, D. what is the relationship between improve- data do tell many important things about (2004). Auditing best practice for effec- ment in project culture, and efficiency introducing project-based management tive product innovation management. improvement? What is the temporal link- in these firms. Regarding the survey Technovation, 24(10), 819–829. age between internal complexity and design, the use of subjective estimates of Damanpour, F. (1987). The adop- external pressure? What kind of factors the introduction of project-based man- tion of technological, administrative, drives the degree of process change and its agement may have its drawbacks. and ancillary innovations: Impact of organizational factors. Journal of impact on local success in project-based Knowing that the sample was dominated Management, 13(4), 675–688. management adoption? What behavioral by very experienced project people, the Damanpour, F. (1996). and attitude changes should be consid- results could have looked different, had Organizational complexity and inno- ered as intermediary impacts of adopting we had access to multiple opinions or vation: Developing and testing multiproject-based management? objective measures in the same firms. To ple contingency models. Management Institutional theory and innova- improve the applicability of the findings, Science, 42(5), 693–716. Damanpour, F., & Evan, W. M. tion diffusion research encourage we have reported the sample characteris- (1984). Organizational innovation studying the role of standardization tics as thoroughly as possible. and performance: The problem of and project management association The validity of the survey and devel- “organizational lag.” Administrative membership with regard to the adop- oped variables could have been improved Science Quarterly, 29, 392–409. tion of innovations. Project manage- by further testing and refinement. With Damanpour, F., Szabat, K. A., & ment research has to some extent the questions and scales used, the relia- Evan, W. M. (1989). The relationship already covered standardization of bility of some variables was slightly between types of innovation and organizational performance. Journal project management, but its link with below the acceptable level of 0.7, the of Management Studies, 26(6), project management maturity and validity of the entire factor structure could 587–601. evolution could be studied further. not be confirmed, and many interesting DiMaggio, P. J., & Powell, W. W. Top management support has areas of innovation adoption remained (1983). The Iron Cage Revisited: already been mentioned as potential uncovered. In this sense, we must consid- Institutional Isomorphism and area for research. Our survey did not er this study as exploratory: we probed Collective Rationality in Organizational cover top management actions and with a set of questions and variables, suc- Fields. American Sociological Review, 48(2), 147–160. practices directly but only in the ceeded in charting important aspects of Fennell, M. L. (1984). Synergy, form of a control variable “official project-based management as an organi- influence, and information in the introduction of project-based man- zational innovation, and opened up are- adoption of administrative innova- agement.” Earlier studies emphasize nas for further research. tions. Academy of Management Journal, the role of top management support 27(1), 113–129. that could be examined also in the References Fennell, M. L., & Alexander, J. A. connection with introducing project- Abrahamson, E. (1991). Managerial (1987). Organizational boundary spanning in institutionalized environ- based management. fads and fashions: The diffusion and rejection of innovations. Academy of ments. Academy of Management Journal, More research is also suggested to Management Review, 16(3), 596–612. 30(3), 456–476. examine the diffusion of project-based Alänge, S., Jacobsson, S., & Guler, I., Guillén, M. F. & Macpherson, management within and across industries. Jarnehammar, A. (1998). Some aspects J. M. (2002). Global competition, institu- Institutional theory and innovation diffu- of an analytical framework for study- tions, and the diffusion of organizational sion literature provide a good basis and ing the diffusion of organizational practices: The international spread of ISO suggestions regarding relevant hypotheses innovations. Technology Analysis and 9000 quality certificates. Administrative and contingency factors. Strategic Management, 10(1), 3–21. Science Quarterly, 47, 207–232.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 95 Hensman, N., Valenta, K., & Jaafari, A. Heterogeneity and transferring practices: Planning, 31, 699–725. (2004). Project management in Australia: Implementing flow manufacturing in mul- Teece, D. J. (1980). The diffusion of State of play and trends. In K. Wikström & tiple plants. Strategic Management Journal, an administrative innovation. Management K. A. Artto (Eds.), Proceedings of the IRNOP 24, 945–959. Science, 26(5), 464–470. VI Project Research Conference (pp. Meyer J. W., & Rowan, B. (1991). Turner, J. R. (1999). Handbook of proj- 292–310). Turku, Finland: Åbo Akademi Rationalized organizations: Formal structure ect based management, improving the University and Helsinki University of as myth and ceremony. In W. W. Powell & P. processes for achieving strategic objectives, Technology. J. DiMaggio (Eds.), The New Institutionalism second edition. Glasgow, UK: McGraw Hill. Ibbs, C. W., & Kwak, Y. H. (2000). in Organizational Analysis (pp. 41–62). The Volkswagen Coaching. (2002). Assessing project management maturity. University of Chicago Press. Report 10/2002. Bremen, Germany: Project Management Journal, 31(1), 32–43. Milosevic, D., & Patanakul, P. (2005). Volkswagen Coaching. Jugdev, K., & Thomas, J. (2002). Standardized project management may Westerveld, E. (2003). The project Project management maturity models: The increase development project success. excellence model®: Linking success criteria silver bullets of competitive advantage? International Journal of Project Management, and success factors. International Journal of Project Management Journal, 33(4), 4–14. 23(3), 181–192. Project Management, 21, 411–418. Kimberly, J. R., & Evanisko, M. J. Packendorff, J. (1995). Inquiring into Westphal, J. D., Gulati, R., & Shortell, (1981). Organizational innovation: The the temporary organization: New direc- S. M. (1996). The institutionalization of influence of individual, organizational, tions for project management research. total quality management: The emergence and contextual factors on hospital adop- Scandinavian Journal of Management, 11(4), of normative TQM adoption and the con- tion of technological and administrative 319–333. sequences for organizational legitimacy innovations. Academy of Management Project Management Institute. (2004). and performance. Academy of Management Journal, 24(4), 689–713. A guide to the project management body of Proceedings, 249–253. Kwak, Y. H., & Ibbs, C. W. (2002). knowledge (PMBOK® guide) (3rd ed.). Westphal, J. D., Gulati, R., & Shortell, Project management process maturity Newtown Square, PA: Project Management S. M. (1997). Customization or conform- (PM)2 model. Journal of Management in Institute. ity? An institutional and network perspec- Engineering, 18(3), 150–154. Powell, T. (1995). Total quality man- tive on the content and consequences of Mahajan, V., Sharma, S., & Bettis, R. A. agement as competitive advantage: A review TQM adoption. Administrative Science (1988). The adoption of the M-form orga- and empirical study. Strategic Management Quarterly, 42, 366–394. nizational structure: A test of imitation Journal, 16, 15–37. White, D., & Fortune, J. (2002). hypothesis. Management Science, 34(10), Shenhar, A. J., Dvir, D., Levy, O., & Current practice in project manage- 1188–1201. Maltz, A. C. (2001). Project success: A mul- ment—An empirical study. International Maritan, C. A., & Brush, T. H. (2003). tidimensional strategic concept. Long Range Journal of Project Management, 20(1), 1–11.

MIIA MARTINSUO, PhD, is a senior lecturer in industrial management at Helsinki University of Technology, Finland. Her teaching, student instruction, and research work focus on the following aspects of project business: multiproject management, innovation management, and organization development. Dr. Martinsuo has more than 10 years of experience in strategy, business process development and management development in international private companies, especially in metal and electrotechnical industry. She has consulted industry, service and public sector organizations as well as managed international research and development projects toward improved organizational efficiency and well-being. She has written and co-authored more than 60 papers, book chapters and books especially in the areas of project portfolio management and organization development. Dr. Martinsuo holds DSc, Lic.Tech., and MSc degrees in industrial engineering and management.

ALI JAAFARI, PhD, is professor and founding president at Asia Pacific International College in Sydney, Australia. As a long time international consultant, author, researcher and educator in project, program and business management Professor Jaafari has wide expertise and professional experience. He has acted as an expert consultant to industry and governments worldwide. Professor Jaafari has conducted courses and seminars for more than 3,000 executives, managers and professionals in Australia, Asia, and Europe. He specializes in graduate education and professional development, and has developed innovative online graduate programs that have won numerous Excellence Awards. Professor Jaafari has, to-date, authored 150 publications in project, program, organizations and business management. Professor Jaafari’s current research efforts are focused on strategic management of projects, programs and organizations, project, program and portfolio management, innovation and change management, information management systems, total quality management and management of risks and liabilities.

JAAKKO KUJALA, PhD, works as a senior researcher in BIT Research Center and lecturer in the Department of Industrial Engineering and Management. He has an extensive industrial background in Metso Corporation in managing both customer delivery projects and internal development projects. He holds an MSc degree in information technology and he received doctoral degree at the Department of Industrial Engineering and Management. He is responsible for courses in the area of project marketing and customer relationship. His main research interests include network perspective to projects, project marketing and project business management. In addition, he is also actively involved in the research in healthcare management.

KARLOS ARTTO, PhD, is professor at the Laboratory of Industrial Management at the Helsinki University of Technology, Finland, where he is responsible for developing research and teaching in the field of project-oriented activities in business. His current research interests in project business cover: (1) management of project-based organizations and strategic management of multiple projects, project portfolios, and programs; (2) management of innovation, technology, R&D, new product development and operational development projects in different organizational contexts; (3) project networks and project delivery chains, and; (4) management of uncertainty and risk in project business.

NICOLE HENSMAN is a doctoral student at the University of Sydney, with project management as her field of research.

96 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Appendix 1: Variables, items included in them, principal component analysis results, and reliability coefficients (Cronbach’s alpha)

Drivers, Reasons for Introducing PM Component 1 Component 2 Component 3

Internal complexity 0.66 Increasing project complexity 0.82 Increasing number of projects 0.80 Time pressure for projects 0.66 0.40 External pressure 0.68 Market /competitive pressure 0.79 Client demands 0.72 Image of modernity 0.71 Internationalization / globalization 0.53

Loadings below 0.3 omitted Changes Degree of process change 0.83 Degree of process change in your department 0.87 Degree of process change in your area 0.84 Degree of process change for you personally 0.81 Depth of PM adoption 0.73 PM culture is widely present at all levels of the hierarchy 0.87 PM is used consistently (=not sporadically) in the company 0.82 Project and line organizations work well together in the company 0.67 Local success of PM introduction 0.89 How successful was PM introduction in your department? 0.90 How successful was PM introduction in your area? 0.88 Loadings below 0.3 omitted Benefits of Introducing PM

Improvement of project culture 0.80 Greater entrepreneurship 0.92 More knowledge management, know-how transfer 0.73 More client satisfaction 0.69 More effective communication 0.67 Efficiency improvement 0.86 Better multiproject coordination -0,88 Improved project control -0,81 Greater project transparency -0,80 Better project performance 0.47 -0,55 Loadings below 0.3 omitted

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 97 A THEORETICAL FRAMEWORK FOR ALIGNING PROJECT MANAGEMENT WITH BUSINESS STRATEGY

DRAGAN Z. MILOSEVIC, PMP, Portland State University SABIN SRIVANNABOON, Portland State University

Introduction ABSTRACT istorically, the world of business has recognized business strategy planning, portfolio management, and project selection as the responsibilities gov- This study addresses two aspects of a H erned by senior managers and project planning and execution processes as topic under-researched in the strategic management literature: the alignment of the activities performed by project managers and their project teams. When these project management and business strat- processes are aligned, the strategic element feeds the portfolio element, the port- egy. Two areas of this alignment were folio element feeds the project management element, and the project manage- studied: (1) The reciprocal influence ment element feeds projects and the team’s execution. But in many cases, these between project management and busi- processes are not aligned; as a result, organizations may fail to tie their projects ness strategy, which we call the nature of the project management/business strat- either to their business strategy or to their portfolio, which may cause them to ter- egy alignment; and (2) the process used minate the project or to continue implementing projects that do not contribute to to align project management and busi- the organization’s goals, thus wasting important organizational resources. In ness strategy. Then an empirically based many instances, organizations treat all projects in the same way, regardless of the theoretical framework, which highlights business strategy that the organization chooses (Pinto & Covin, 1989; Shenhar, the impact of business strategy on project management—and the impact of 2001). When the organization’s business strategy is translated into project-level project management on business strate- goals, its professional uniqueness—such as speed to market, superior product gy—as well as the mechanisms used to quality, among other factors—may dissolve. By understanding the challenges strengthen these alignments, was devel- involved in aligning project management and business strategy, practitioners can oped. This study expands on the previ- effectively manage their projects in today’s competitive environment. ous, mostly anecdotal work, by using a rigorous theoretical approach to develop Scant, however, is the empirical literature on aligning project management the proposed framework. This framework and business strategy. This study, however, addresses lack of information by is contingent upon the type of business exploring two aspects of aligning project management and business strategy: strategy—simple to understand and 1. A two-way influence between project management and business strategy, use—developed through numerous proj- one suggesting the nature of the alignment between project management ects that are typology-free and not restricted to any particular business and business strategy. strategy typology, through projects that 2. A process used for aligning project management and business strategy. are empirically based on real-world data.

Keywords: qualitative research; project man- We have developed an empirically based theoretical framework that shows the agers; pragmatic knowledge; skills development impact of business strategy on project management—as well as the effect of proj- ©2006 by the Project Management Institute ect management on business strategy—and discusses the mechanisms used to Vol. 37, No. 3, 98-110, ISSN 8756-9728/03 strengthen that alignment. We define this framework in regard to a set of well developed concepts related to each other by statements of interrelationships, statements that include an integrated structure that can be used to describe phenomena in a manner similar to the concept of theory defined by Strauss and Corbin

98 AUGUST 2006 PROJECT MANAGEMENT JOURNAL (1998). We also refer to the alignment between project man- • Differentiation: Organizations pursuing a differentia- agement and business strategy in relation to the compatible tion strategy seek to position themselves in the mar- priorities between an organization’s project management ketplace with a distinct identity that satisfies the practice and its business strategy. desires of their customers (e.g., fast time-to-market, superior quality and service, innovative features). This Theoretical Background differentiation allows the organization to charge a To develop a theoretical framework for aligning project premium price (Porter, 1980). management with business strategy, we examined multiple • Best-cost: Under certain conditions, many researchers streams of related literature, streams that include business argue that a combination of strategies may be the best strategy and its typology (i.e., understanding the definitions way of creating a sustainable competitive advantage of business-level strategy and the conceptual basis of differ- (e.g., Hill, 1988; Miller & Friesen, 1986; Phillips, ent strategic types), project management (i.e., identifying Chang, & Buzzell, 1983; White, 1986). In particular, project management elements that should be aligned with organizations may more effectively create a sustain- business strategy), and alignment literature (i.e., studying able competitive advantage when they combine cost previous and recent alignment research to identify what has leadership and differentiation, when they provide been done and what is missing). low-cost products and address customer values (fast time-to-market, superior product quality, etc.). Business Strategy and Business Strategy Typologies Though the definitions of business strategy vary, these—in Project Management general—do focus on how to better deal with competition Project management is a specialized form of management, (Tse & Olsen, 1999) by means of creating competitive similar to other functional strategies, that is used to accom- advantages (Hamel & Prahalad, 1989), advantages that pro- plish a series of business goals, strategies, and work tasks vide organizations with the benefits that will sustain them within a well-defined schedule and budget. The essence of when attracting customers and defending themselves project management is to support the execution of an orga- against competitive forces (Thompson & Strickland, 1995). nization’s competitive strategy to deliver a desired outcome Although the literature discusses multiple business-strategy (i.e., fast time-to-market, high quality, low-cost products) typologies, organizations should only consider those that (Milosevic, 2003). As opposed to the traditional stereotype, align with their project management practice and their busi- the recent literature recognizes project management as a key ness strategy, e.g., Miles and Snow’s typology (1978), business process (Jamieson & Morris, 2004). This view Porter’s generic strategies (1980), Treacy and Wiersema’s defines an organization as the process rather than the tradi- typology (1995). In this paper, we present only one, Porter’s tional functional or matrix form and describes project man- generic strategies, using it as the foundation for aligning agement as one of the key business processes that enable project management and business strategy (see the Research companies to implement value delivery systems. Therefore, Design section for information about our reasoning). when organizations link their projects to their business Porter (1980) claimed that to achieve a sustainable strategy, they are better able to accomplish their organiza- competitive advantage, an organization must reinforce its tional goals. Shenhar’s strategic project leadership (SPL) chosen strategies. Depending on the scope, there are three framework (1999) identifies the project management ele- generic strategies that can result: cost leadership, differentia- ments that organizations should align with business strate- tion, and focus. According to Porter, generic strategies— gy, elements such as project strategy, organization, process, when an organization chooses only one—provides the tools, metrics, and culture. (For this paper, we have adapted organization with the ability to achieve competitive advan- the SPL framework, adding metrics and changing project spir- tages and outperform their competitors. However, if an it to project culture.) organization pursues more than one generic strategy, it will perform below its capability. Porter referred to the latter type Alignment Literature of organization as stuck-in-the-middle. Despite this, the pro- Research in the literature has examined the idea of align- liferation of global competition is compelling more organi- ment in various management areas. For example, numerous zations to focus on a single combination of generic studies have discussed the alignment between tasks, poli- strategies (Harrison & St. John, 1998). Many researchers and cies, and practices (e.g., Boyer & McDermott, 1999; Kathuria practitioners (e.g., Hill, 1988; Miller & Friesen, 1986; & Davis, 2001); others have emphasized the relationship Phillips, Chang, & Buzzell, 1983; White, 1986) refer to this between alignment and performance in regards to organiza- combination as the best-cost strategy. In this paper, we used tional hierarchy: corporate, business, and function (e.g., three of the previously mentioned business strategies for our Papke-Shields & Malhotra, 2001; Youndt, Snell, Dean, & analysis, each of which we define as follows: Lepak, 1996). The literature frequently mentions research • Cost leadership: Organizations pursuing a cost leader- and development (R&D), production, human resources, ship strategy seek to gain competitive advantage and and information technology—among others—as functional increase market share by being the lowest cost pro- strategies and uses these as the variables to examine align- ducers in the industry (Porter, 1980). ment in relation to the business strategy. Because project

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 99 management is similar to these functional strategies, it too order to obtain information from different perspectives should be aligned with the business strategy (Harrison, (Boynton & Zmud, 1984). In addition to the interviews, we 1992). However, the traditional literature on aligning proj- reviewed related documents—meeting minutes, project ect management with the business strategy is vague: Most descriptions, risk logs—to triangulate and validate our findings. studies link the business strategy with project management In this study, we determined a case study to be a study through project selection, viewing it as part of the alignment of a project in a distinguishable business unit, where a proj- process (e.g., Baker, 1974; Bard, Balachandra & Kaufmann, ect is being executed. To select the reviewed cases (compa- 1988; Cooper, Edgett & Kleinschmidt, 1998a; Englund & nies, projects, and participants), we defined multiple criteria Graham, 1999; Hartman, 2000). Added to this is project and identified the cases most relevant to such criteria as the- portfolio management (PPM and also called pipeline man- oretical sampling and project frame of reference (projects agement), another concept suggested in the literature to completed in at least six month or under) as well as the proj- ensure the strategic alignment of project management and ect management experience of the participants (at least business strategy (Turner & Simister, 2000). Cooper, Edgett, three years). and Kleinschmidt (1998b) defined PPM as a dynamic deci- We then classified these projects into different types, sion-making process through which an organization can including strategic projects (creating strategic positions in update and revise its list of active projects. The organiza- markets and businesses), extension projects (improving or tion’s choice of business strategy is what drives their PPM upgrading an existing product), utility projects (acquiring process, the major purposes of which are to select and pri- and installing new equipment or software, implementing oritize projects (Cooper et al., 1998b), balance projects new methods or new processes, reorganization, reengineer- (Archer & Ghasemzadeh, 1999; Cooper et al., 1998b), align ing), and R&D projects (exploring future ideas, no specific projects with the business strategy (Cooper et al., 1998b), product in mind). These projects were also categorized in manage rough-cut resource capacity (Harris & McKay, 1996; regards to external customers (external contract or con- Wheelwright & Clark, 1992), and articulate empowerment sumers), internal customers (internal users or another boundaries for project and functional management (Harris department), or both. We also evaluated each project in rela- & McKay, 1996). tion to such success dimensions as project efficiency, impact Only recently have researchers started to explore the on the customer, direct organizational success, and team alignment of project management more thoroughly (e.g., leader and team spirit. Artto & Dietrich, 2004; Jamieson & Morris, 2004; Papke- After each interview (phase 2: data analysis), we tran- Shields & Malhotra, 2001; Srivannaboon & Milosevic, scribed the conversation and coded it. We then wrote case 2004). For example, Jamieson and Morris (2004) suggested studies—25–30 pages per case—about our interviews and that most of the components comprising the strategic plan- study of the related documents. We sent these cases to the ning process—internal analysis, organizational structures, companies to verify the accuracy of our transcriptions so as control systems—have strong links to project management to enhance the validity of the research. We then performed processes and activities. As a result, these strongly influence within-case, cross-case, and content analyses. Altogether, we an organization’s intended business strategies. Similarly, studied eight cases (Cases A to H) in seven organizations, a Artto and Dietrich (2004) suggested that an important man- study that involved nine projects of differing size, type, and agerial challenge involved in aligning project management complexity (42 interviews). During phase 3, we engaged a and business strategy is encouraging individuals to partici- panel of five experts—from academia (three professors) and pate in using emerging strategies to create new ideas and industry (two practitioners)—to validate the essential find- renew existing strategies. These studies suggest a need for ings. These experts generally agreed on the findings; they more research in this area; none, however, explicitly talks also contributed views, which we integrated into our find- about the process used to align project management and ings to sharpen our theoretical framework. business strategy cohesively and comprehensively. For each case study, we employed a self-typing method (Conant, Mokwa, & Varadarajan, 1990) to classify the busi- Research Design ness strategy, one based on Porter’s generic strategies To complete this study, we integrated two overlapping (1980), which we used to illustrate the impact of the busi- research phases: data gathering and data analysis. During ness strategy types on the composition of project manage- data gathering (phase 1), we conducted a literature review ment elements. We chose Porter’s generic strategies to so as to understand the general research on aligning project classify business strategies types because of the following: management and business strategy. In parallel with our lit- • Porter’s generic strategies are well accepted and oper- erature review, we researched case-studies over a 10-month ationalized in the literature (e.g., Harrison & St. John, period, studying the nature of alignment in market-leading 1998; Kim & Lim, 1988; Miller & Dess, 1993; organizations through semi-structured interviews (ranging Reitsperger, 1993; Veliyath, 2000). from 60 to 120 minutes per interview) with individuals • Porter’s generic strategies focus on the strategic posi- holding key organizational positions, individuals such as tioning dimension of the business strategy (Kald, senior managers, project managers, assistant project man- Nilsson, & Rapp, 2000), the underlying way in which an agers, and team members—as well as a few customers—in organization relates to its product, where differentiation

100 AUGUST 2006 PROJECT MANAGEMENT JOURNAL (i.e., quality, time-to-market), cost, and a combina- competitive attributes associated with a customer tion of both are often addressed as a project’s major focus. For example, the structures of projects AS, AUS, objectives, constraints, and requirements. and BS were relatively flexible in order to help them achieve their desired outcome (speed, quality, etc.). In studying the nature of alignment, we adapted the ele- 3. Project process: Project process is relatively flexible, ments of project management from Shenhar’s SPL frame- when compared to other projects in this study, and work (1999), elements such as project strategy, mandated by its competitive attributes of customer organization, process, tools, metrics, and culture. We adopt- focus. For example, we observed the overlapped and ed this framework because it is well-publicized and tested. combined phases in time-to-market differentiation Our study comprised two differentiation strategy companies (Case A); we found that the iterative phases ensure (Cases A and B), one cost leadership company (Case H), and the best quality in quality differentiation (Case B). five best-cost companies (Cases C, D, E, F, and G). We coded 4. Project tools: Time-to-market differentiation focuses on the examined project as xS or xUS, where x represents a case, scheduling tools, wherein cost tools are more flexible S represents a successful project in that case (projects AS, BS, than scheduling tools. Quality differentiation focuses CS, etc.), and US represents an unsuccessful project in that on quality control tools, wherein schedule and cost case (project AUS). tools are more flexible than quality control tools. 5. Project metrics: Project performance measures are Results directed by the competitive attributes determined by In this section, we divide our findings into two subsections: the differentiation strategy (e.g., the ability of projects The nature of the alignment and the process used for the to meet the schedule, feature sets, quality, and finan- alignment. cial expectations). Similar to project tools, time-to- market differentiation focuses on scheduling metrics, The Nature of Aligning Project Management and Business Strategy wherein cost metrics are more flexible than schedul- First, we analyze the patterns of each of Porter’s generic ing metrics. Quality differentiation focuses on quality strategies in relation to each project management element. control metrics, wherein schedule and cost metrics are Then, we propose six propositions—one for each project more flexible than quality control metrics. management element—at the end of each generic strategy. 6. Project culture: A project culture of time-to-market dif- Propositions for differentiation are represented as D, propo- ferentiation is built around the schedule focus where sitions for cost leadership as C, and propositions for best-cost projects must be finished at the earliest time possible. as BC. We also use content analysis to compare cases and Therefore, a rapidly changing environment is com- develop generic propositions (P) that address individual mon; project teams are taking risks and proactively project management elements without reference to any spe- accelerating the project cycle time; as a result, senior cific type of business strategy. Lastly, we explain the recipro- managers reward their speed. Similarly, product qual- cal relationship between project management and business ity is driving the project’s culture of quality differen- strategy, discussing these as the emergent strategic feedback tiation; thus, project teams communicate openly and adapting business strategy. extensively to ensure they achieve a high level of product quality. Such efforts are usually rewarded by Patterns in Project Management Elements for Differentiation senior managers. Business Strategy From these patterns we have developed six proposi- 1. Project strategy: General rules to guide the behavior tions, one for each project management element: (strategic focus) of the project teams—designed to On the basis of its competitive attributes, the differenti- help accomplish the goals of differentiation—are ation strategy generally drives the focus and content of the rooted in competitive attributes (fast time-to-market, following: superior product quality). For example, the teams Proposition D1: project strategy implementing projects AS and AUS were directed by Proposition D2: project organization their senior managers to drop some product features, Proposition D3: project process if necessary, in trade-off situations so as to maintain Proposition D4: project tools the project’s time-to-market focus as mandated by Proposition D5: project metrics differentiation. Similarly, senior managers guiding Proposition D6: project culture the team realized project BS would delay the project’s schedule by three months; as a result, they fixed the Patterns in Project Management Elements for Cost Leadership functionalities so as to retain the focus and content Business Strategy needed to achieve superior product quality, a com- 1. Project strategy: Project strategy is driven by cost lead- petitive attribute of differentiation. ership with the purpose of creating competitive 2. Project organization: The project organization tends to advantage through a cost reduction (e.g., process possess a high degree of flexibility when compared to improvement), which may or may not lead to under- other projects in this study; it is aiming to achieve the pricing the competition. Schedule is important

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 101 because it helps the cost leadership company save 3. Project process: The project process is standardized and money if the project finishes on time. built on templates. Every project follows the same 2. Project organization: The structure of an organization’s steps with a keen emphasis on achieving the best cost leadership strategy is flexible, when it is compared quality, innovative features, or desired science at the to other projects in this study, enough to adapt to a lot minimum cost, as in projects CS, DS, ES, FS, and GS. of change through process improvement so as to attain 4. Project tools: Customer voice is crucial for hitting the its ultimate goal of saving costs. customer’s required quality level and innovative fea- 3. Project process: The project process of cost leadership ture level in addition to the cost estimates and base- strategy is highly standardized and built on tem- lines. Other tools—for schedule, scope, and risk—are plates. The observed project followed the generic also used throughout the project life cycle. steps and procedures created by the organization (or 5. Project metrics: Similar to project tools, project business unit). Because standardization reduces vari- progress is measured by the ability of projects to meet ation and cost, the idea was that every project follows or exceed the specification of the expected products the same steps. while still maintaining or minimizing expected costs. 4. Project tools: Schedule tools are important because Quality assurance, cost, and schedule metrics are these help projects finish on time, thus helping dominant, important, and used throughout the proj- increase cost savings. Cost estimates and cost baselines ect life cycle. are required; Gantt charts are often used as a visual dis- 6. Project culture: To maintain a high level of product play of the project schedule. quality with a minimum cost, the examined project 5. Project metrics: Schedule metrics are used as techniques culture of the best-cost strategy included open com- for tracking projects; by meeting target dates, organiza- munication, intensive preparation, trade-off consid- tions can save money. Cost-saving, or net present value erations, and rewarding project teams for product (NPV), is the ultimate measure of project success. quality and cost efficiency. 6. Project culture: Team spirit is cost-centric, focusing on From these patterns, we have outlined six propositions, cost reduction goals and getting the job done. Some one for each project management element: observed attributes include open communication, On the basis of its competitive attributes, the best-cost flexibility, and cost efficiency. strategy generally drives the focus and content of the following: Proposition BC1: project strategy From these patterns we have created six propositions, Proposition BC2: project organization one for each project management element: Proposition BC3: project process On the basis of its competitive attributes, the cost lead- Proposition BC4: project tools ership strategy generally drives the focus and content of the Proposition BC5: project metrics following: Proposition BC6: project culture Proposition C1: project strategy Proposition C2: project organization Figure 1 summarizes the configuration of project man- Proposition C3: project process agement elements, as influenced by each type of Porter’s Proposition C4: project tools business strategies that we have previously discussed. Proposition C5: project metrics Proposition C6: project culture Patterns in Project Management Elements for Porter’s Generic Strategies Patterns in Project Management Elements for Best-Cost The propositions presented in the previous section are stat- Business Strategy ed in a way that is specific to Porter’s generic strategies. To 1. Project strategy: The focus and content of project strat- generalize these even more, we used a content analysis egy are driven by the combination of its competitive process to develop generic propositions that address indi- attributes (e.g., quality, innovative, customization, vidual project management elements without reference to science) determined by best-cost strategy and cost. any specific type of business strategy. The content analysis For example, project strategies of CS, DS, ES, FS, and process searched for what the three strategic types have in GS were developed to balance customer needs (e.g., common in regards to how the business strategies dictate quality, innovation, science) and project resources. the configuration of project management elements. The The key is to find the level of the differentiation at a comparison of the propositions that describe how business reasonable cost. strategy types of differentiation, cost leadership, and best- 2. Project organization: Project organization is fairly flex- cost impact the project management elements revealed pat- ible, when compared to other projects in this study, terns: All three strategic types influence project management and often involves different functions with the aim of elements through the competitive attributes that were cho- ensuring the best quality, innovative features, or sen as a basis of competition for individual strategic types. desired science, and accomplishing this while From the patterns previously outlined, we suggest six decreasing project cost. propositions, one for each project management element:

102 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Degree of Differentiation Low High

(1) Differentiation Strategy (Examples: Time-to-market or quality differentiation) Emphasis is placed on: • (Strategy) Schedule or quality project success measure • (Organization) A flexible structure to facilitate project speed or product quality • (Process) A flexible process to speed up projects or maximize product quality • (Tools and Metrics) Schedule- or quality-oriented tools and metrics • (Culture) Rewarding time-to-market speed or quality Cost (2) Cost Leadership Strategy (3) Best-Cost Strategy (Example: Process improvement) (Example: Quality/cost) Emphasis is placed on: Emphasis is placed on: • (Strategy) Cost-efficiency project success measures • (Strategy) Quality and cost project success • (Organization) A flexible structure to adapt to measures changes in process improvement • (Organization) A flexible structure to ensure the best Low High • (Process) A highly standardized and built-on product quality at the minimum cost template process • (Process) A standardized but flexible process • (Tools and Metrics) Cost- and schedule-driven tools • (Tools and Metrics) Quality/cost-oriented tools and and metrics metrics • (Culture) Cost-conscious culture • (Culture) Rewarding quality/cost culture

Figure 1: Summary of project management configuration per Porter’s generic strategies

The competitive attributes of the business strategy An explicit example of this relationship is Project AUS and drive the focus and content of the following: its business strategy. This project’s failure is related to the win- Proposition 1: project strategy dow of opportunity. Although the project was initially aligned Proposition 2: project organization with the organization’s business strategy, the product that Proposition 3: project process resulted from the project was released after the market had Proposition 4: project tools shifted and customers began looking for a more complex Proposition 5: project metrics product. This project also failed because the project team did Proposition 6: project culture not appropriately validate the product definition (as part of the project’s strategy) with the key customers throughout its Reciprocal Relationship of Project Management and life cycle. As a result, Project AUS failed because of inefficient Business Strategy stage gate reviews that lacked the feedback necessary to detect Interestingly, we found cases where project management significant threats, such as a market shift. The company, how- elements not only support but also impact business ever, later adjusted its stage gate reviews to cover market shifts strategy. We call this relationship the reciprocal relation- as a measure to prevent such failure from repeating. ship of project management and business strategy. This rela- This example implies that in order to ensure project per- tionship occurs when companies obtain from their formance, project managers must realign the project strategy, projects information about the ways they adapt their the organization and its culture, and the processes, tools, met- business strategy, a process that Mintzberg (1994) ref- rics of realizing projects with a project’s progress. Another ered to as an emergent strategy approach, one also known proposition concerning the reciprocal relationship between as the redirection of projects. project management and business strategy involves the oper-

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 103 ating conditions of reviewed projects, which are revealed at process, tools, metrics, and culture) is to accomplish the time- stage gate reviews. Results of stage gate reviews may impact the to-market competitive attribute. This study defines this focus business’s strategies and its competitive attributes because of as schedule-driven (see Figure 2). The content or configuration environmental changes (also known as the emergent of project management elements (strategy, organization, approach [Mintzberg, 1994]). process, tools, metrics, and culture) is also tailored to support Proposition 7: Project management elements may this schedule-driven focus. For example, in case study A, the impact business strategy, as based on the operating condi- configuration of the project strategy (P1) was tailored to sup- tions of reviewed projects. port its schedule focus; the time-to-market competitive attribute adopts a strategic focus that allows project managers to A Theoretical Framework: Nature of the Project Management- ignore cost and product features in making trade-off decisions Business Strategy Alignment in order to attain time-to-market goals. The project process To construct a theoretical framework for the configuration of (P3) is similarly tailored to deliver a time-to-market competi- project management elements, one influenced by business tive advantage by overlapping or combining process phases, strategy, we used the seven propositions previously outlined to milestones, and activities. At the same time, operating condi- connect business strategy and each project management ele- tions detected from stage gates (P7) help to redirect projects, if ment (P1 to P6) and vice versa (P7). The nature of the project there is any change that might threaten the success of the proj- management/business strategy alignment is depicted in the ects. There are infinite combinations of competitive attributes theoretical framework in Figure 2 as the impacting nature, as that companies can use as sources of advantage to compete that which addresses the relationship between the competitive with their rivals. There are also unlimited alternatives for tai- attributes of business strategy and the focus and content of loring project management elements to support these compet- project management elements. itive attributes. Propositions 1 to 6 demonstrate how the A company (business unit) makes its strategic choice by competitive attributes of business strategy configure the indi- selecting competitive attributes that are advantageous (e.g., vidual project management elements. This should lead to one time-to-market, quality, cost, and features). These attributes single and generic proposition that describes the interaction of are used to drive the different ways that projects are managed the business strategy and project management elements: in terms of their foci and contents. For example, if the com- Generic Proposition: The competitive attributes of the petitive attribute of time-to-market is chosen, the focus or pri- business strategy drive the focus and content of project man- ority of project management elements (strategy, organization, agement elements.

Business Project Management Strategy Impacting Nature Elements

Competitive PM Elements PM Elements Attributes Focus Contents P1* Strategy P2* Tailored to Support Org. Time-to-Market Schedule-Driven Schedule-Driven Focus The P3* Competitive Process Attributes of Business Tailored to Support Strategy Quality Quality-Driven P4* Quality-Driven Focus Tools

P5* Metrics Tailored to Support Cost Reduction Cost-Driven Cost-Driven Focus P6* Culture

Tailored to Support Feature Feature-Driven Feature-Driven Focus

P7* *Propositions

Figure 2: A theoretical framework for the nature of the alignment

104 AUGUST 2006 PROJECT MANAGEMENT JOURNAL The Process Used for the Project Management/Business Level 2—Mediating Process at the Project Level Strategy Alignment Once organizations select projects into their portfolio, they In this subsection, we analyze the patterns of the processes further plan the details and execute these throughout the proj- used by the companies we surveyed to align project man- ect life-cycle phases. We refer to these mechanisms so as to agement and business strategy. In doing so, we discuss the ensure the proper alignment during the project life cycle as similarities and the dissimilarities across all cases in order to the mediating processes at the project level, which can be clas- generate a theoretical framework of the processes that sified into the planning process and the monitoring process. organizations use to ensure proper alignment. We per- In the planning process, we found that the companies formed a content analysis to compare these cases and iden- used varying mechanisms to ensure proper alignment. The tify the patterns of the alignment processes used across these most explicit planning mechanism used was in Case C: This cases. The pattern we found revealed that organizations company required that project managers identify the align- could divide the mechanisms used to align projects with ment link of their project plans and the goals in their strategic business strategies into three levels: the strategic, the tactical, plan. This was accomplished through product definition and and the corrective emergent strategic feedback. Each levels project definition, by linking these with the business goals out- contained distinct mechanisms to achieve alignment. Table 1 lined in the strategic plan. In the other cases, this was implicit- summarizes the alignment process of different cases and ly accomplished through the development of the project plan, patterns upon these levels. as based upon the objectives of the projects and the reason why these existed, such as achieving business goals. Level 1—Mediating Process at the Strategic Level We found that as projects progress, most companies use The general steps of the alignment process begin at the strate- common mechanisms to ensure these are properly aligned gic level, where the long-term business goals are defined and during execution, using mechanisms such as project metrics, business directions are determined through a strategic plan, internal coordination mechanisms (i.e., project manage- through what Mintzberg (1994) called an intended strategy. ment office involvement and internal sign-off), customer We found that every sample company had a strategic plan; involvement (sign-off), and stage gates. This last item, stage some used a formal plan, some used an informal one. In all gates, is so important that we have separated it from this sec- but two cases, these plans were developed to reflect a three- tion to explain it separately as the mediating process at the year planning horizon. One exception was Case B, which at emergent strategic feedback level. the time of our interview was a short-term plan (one-year horizon) that the company was actively expanding to a three- Level 3—Mediating Process at the Emergent Strategic year range. The other exception was Case G, which used an Feedback Level informal plan due to the nature of its business (construc- Stage gates are points in the project life cycle where projects tion). In some cases, roadmaps were included in the strategic transition from stage to stage. The gates represent filters for plan as the guidance for the company’s (or department’s) project status and provide project teams with the opportu- future interests, such as a product roadmap (Cases A and B) nity to realign the project to the requirements set by the and an information technology roadmap (Case D). project owner. In the sample companies, we observed such We also observed that the sample companies used a stage gates as milestone reviews for evaluating the project project portfolio process—again, some used a formal status (time, cost, performance). An exception to this obser- process, others used an informal one—as a mechanism for vation is Case A. This company covered staffing level and selecting the most valuable projects that would contribute to market shift considerations as additional concerns. When a the organization’s goals. To select such projects, and make project fails to meet a stage gate’s requirements (i.e., when them part of the portfolio, many companies used matched the project is misaligned), the project team must adjust the their strategic goals with the project’s contribution, with its project (if the owner has not killed the project), in accor- strategic fit. In several cases, the term project portfolio was dance with the operating conditions of the project. not recognized, but its project selection and prioritization In certain instances where the operating conditions of functions were employed (Cases B, E, F, G, and H). In addi- the project reveals significant changes resulting from inter- tion, two cases recognized the term project portfolio, but it was nal or external factors, revealing factors that may affect the still an informal process (Cases C and D). Only Case A had overall success of the project if the project manager fails to a formal project portfolio management process and semi- manage the changes, the operating conditions will impact annual portfolio reviews, one that included such functions as the deployment of the business strategy by changing the pri- project selection and prioritization, risk balance, strategic orities under which the project is managed. For example, we alignment, and capacity management. Table 2 summarizes found that one of the examined projects in Case A was con- these project portfolio processes and functions used by the sidered an unsuccessful project by its project team and the sample companies. company’s upper management, even though the was initial- In general, the mechanisms to ensure the alignment ly well aligned with the company’s business strategy. Part of process at this level are what we refer to as “the mediating the reason for this perceived failure was that the project was processes at the strategic level,” which include a strategic plan committed to the wrong set of customers, which led to a and project portfolio management. poor product definition of the overall market. By the time

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 105 Case Strategic Level of the Tactical Level of the Corrective Emergent Strategic Alignment Alignment Feedback Level of the Alignment

Case A • Three-year strategic plan (e.g., • Dashboard • Semi-yearly project portfolio product map, technology • Flexibility matrix reviews roadmap, business model) • Bounding box • Gate reviews of individual • Project portfolio projects (market shift, staffing • Project life-cycle phases - Strategic table level, project status)

Case B • One-year strategic plan • Project metrics and monthly • Gate reviews of individual projects (four-stage product marketing, status report (project status) product roadmap) • Project life-cycle phases

Case C • Three-year strategic information plan • Project metrics and monthly • Gate reviews of individual projects • Information technology activity status report (project status) management matrix • Project life-cycle phases • Informal project portfolio

Case D • Three-year strategic information plan • Project metrics and monthly • Gate reviews of individual projects • Roadmap charts status report (project status) • Alignment charts • Project life-cycle phases • Informal project portfolio

Case E • Three-year strategic plan (goals, • Project metrics and monthly • Gate reviews of individual projects recommendations and reviews) status report (project status) • Project life-cycle phases

Case F • Three-year strategic plan (goals, • Project metrics and monthly • Gate reviews of individual projects recommendations and reviews) status report (project status) • Project life-cycle phases

Case G • Informal strategic plan • Project metrics and monthly • Gate reviews of individual projects status report (project status) • Project life-cycle phases

Case H • Three-year strategic plan • Project list • Semi-yearly dashboard reviews • Dashboard • Project metrics and monthly • Gate reviews of individual projects status report (project status) • Project life-cycle phases

Patterns Level 1: Mediating Process at Level 2: Mediating Process at Level 3: Mediating Process at the the Strategic Level the Project Level Emergent Strategic Feedback Level

Strategic planning and project Processes during project planning Stage gates are the mechanisms to portfolio management processes are and execution are the mechanisms ensure proper alignment. the major mechanisms to ensure the to ensure proper alignment. proper alignment.

Table 1: The alignment process and patterns across all cases

106 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Case Formality Used Project Portfolio Process Functions

Case A Formal and recognized Project selection and prioritization, risk balance, strategic alignment, and capacity management

Case B Informal and not recognized (the Project selection term is not used)

Case C Informal but recognized Project selection

Case D Informal but recognized Project selection and prioritization Cases E & F Informal and not recognized (the Project selection term is not used)

Case G Informal and not recognized (the Project selection and prioritization, and risk balance term is not used)

Case H Informal and not recognized (the Project selection and prioritization term is not used)

Table 2: Project portfolio process the project was finished, the operating conditions of the A Theoretical Framework: Process for Project project had changed (the market had shifted), and there Management/Business Strategy Alignment was no longer a place for the product developed through To further develop the theoretical framework we proposed in this project. In this case, the stage gate failed to provide the section titled Nature of the Project Management/Business the organization with the information it needed to realign Strategy Alignment, we combine propositions and mediating its process of managing the project to meet those changes. processes into a single framework, as is shown in Figure 3. The Once the problem was identified at a subsequent stage propositions are used to connect business strategy and each gate, the project team should have adjusted the product’s project management element through statements of relation- definition (as part of the project strategy). Unfortunately, ships (a two-way influence). Mediating processes are mecha- the project team failed to identify in a timely manner the nisms that organizations use to align project management changes that were necessary to save this project. As a and business strategy. For the sake of illustrating the process- result, the team was not able to react to those changes es in general, we have used the traditional phases of the proj- effectively. To accommodate for this unsuccessful effort, ect life cycle, including conception, planning, execution, and the company later adjusted its stage gate reviews to cover closing. Each company, however, uses different project life- market shift considerations. cycle phases, selecting those that are most relevant to their The mechanism previously explained is a feedback industry, company culture, and other significant issues. loop that emerges during project execution. It is a result It is the competitive attributes of the business strategy that is not planned or intended but that emerges from a that drive the focus and the content of the project manage- stream of managerial decisions through time, throughout ment elements. The propositions we have outlined in the what Mintzberg (1994) calls the emergent approach. In framework describe the interrelationships between project other words, the operating conditions of reviewed projects management elements and business strategy. To establish are expected to support the company’s business strategies and maintain the processes used to align project manage- by helping it adapt the business strategy and its competi- ment elements and business strategy, we suggest that organ- tive attributes to environmental changes. izations use mediating processes—strategic planning and Operating conditions refers to the actual conditions of project portfolio management—at the strategic level to project implementation, which may be equal to those interpret their business strategy in the context of project assumed in the project-planning phase. These may also management. Organizations initiate and select projects for differ from those assumed during planning as a conse- their project portfolio to fulfill business needs; they then quence of environmental changes in the marketplace. implement a standard life cycle that includes project plan- These changing business and project conditions can be ning and project monitoring (the primary mediating revealed during the stage gate reviews as well as any phase processes at the project level) to ensure the quality of the of development. Therefore, a combination of intended and alignment between project management elements and busi- emergent strategies is needed to align project management ness strategy. One of the major control mechanisms organi- and business strategy. zations use to ensure that their projects align with their

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 107 Business Mediating Processes PM Elements Strategy (Focus and Content) Strategic Level Project Level P1* PLC Strategy Conceptual P2* Conceptual Org. Strategic Conceptual Planning P3* Planning Process Desired Products/Services Planning Planning P4* Competitive Tools Attributes of High Level Analysis Business Execution Strategy Execution P5* Projects Are Selected Execution Metrics P6* Project Portfolio Closing Culture Closing Management Closing Portfolio Review

Rejected/Killed

P7*

Emergent Strategic Feedback Level

*Propositions Represents Stage Gates Represents a Feedback Loop (emergent approach) when the project is required to change or is rejected at the stage gates

Figure 3: A theoretical framework for aligning project management with business strategy

expectations as the project progresses from one project theoretical framework, as suggested by Dubin (1978), which phase to the next is the stage gate. This mediating process includes units/variables, laws of their interaction, system provides strategic feedback that can lead to what Mintzberg boundaries, and propositions. (1994) calls emergent strategy. • Units/ variables: The variables or units of analysis in the framework consist of two major elements: project Discussion management elements (strategy, organization, process, In this study, we explained an inductive logic process—from tools, metrics and culture) and business strategies (dif- specific to general practices—as a means to derive our ferentiation, cost leadership, and best-cost). propositions. The general process of developing these • Laws of their interaction: The interaction of variables in propositions was based on case study research, which heav- the framework can be seen as a two-way influence ily used within-case, cross-case, and content analyses. We between project management elements and business also developed detailed propositions for Porter’s generic strategy, one that is perceivable through a formal or an strategies, which we generalized into typology-free proposi- informal alignment process by translating business tions. We then developed a single proposition suggesting a needs into project actions and using project operating most generic relationship between project management ele- conditions to more effectively deploy business strategy. ments and business strategy. • System boundaries: The boundary of the framework is Similarly, we used inductive logic to develop an the organizational business units or departments sup- overview describing the mediating processes at different lev- porting them. The project management/business els. Our general process was based on our case study strategy alignment occurs within this boundary. research and used within-case and cross-case analyses. The • Propositions: Seven propositions of the framework framework resulting from this analysis explains the align- are derived from the content analysis of multiple ment process at the strategic level, the project level, and the cases. The propositions explain the unique interac- corrective emergent feedback level. tions of each project management element with the Our framework satisfies the major characteristics for a business strategies.

108 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Our study expands on previous, mostly anecdotal work needed is a large sample study that focuses on the quantita- by incorporating a rigorous theoretical approach into the tive correlations of various strategy types and project man- proposed framework. Although Jamieson and Morris (2004) agement elements. The point here is to find which strategies identified strategic planning, portfolio management, and need which project elements to contribute to project success. emergent approach as important steps in the alignment process, with information that supports this research, they References did not provide a framework and did not position their Archer, N., & Ghasemzadeh, F. (1999). An integrated research as a set of case studies or as a theoretical foundation framework for project portfolio selection. International for alignment. Furthermore, Turner and Simister (2000) Journal of Project Management, 17(4), 207–216. argued, conceptually and without an empirical validation, Artto, K. A., & Dietrich, P. H. (2004). Strategic business that portfolio management is an important step in aligning management through multiple projects. In P. W. G. Morris & projects with the business strategy. In comparison with the J. K. Pinto (Eds.), The Wiley guide to managing projects (pp. existing literature, our framework contributes three elements: 144–176). Hoboken, NJ: John Wiley & Sons, Inc. • Comprehensive: This framework includes—and Baker, N. R. (1974). R&D project selection models: An relates—all levels of participants (executives, middle assessment. IEEE Transactions on Engineering Management, managers, project managers, team members, cus- 21(4), 165–170. tomers), different levels of management processes Bard, J. F., Balachandra, R., & Kaufmann, P. E. (1988). (strategic, tactical, operational), and variables (project An interactive approach to R&D project selection and termi- management elements, business strategy). It integrates nation. IEEE Transactions on Engineering Management, 35(3), these into a coherent structured set of relationships 139–146. based on propositions that describe the phenomenon Boyer, K. K., & McDermott, C. (1999). Strategic consen- of the project management/business strategy align- sus in operations strategy. Journal of Operations Management, ment in different situations. 17(3), 289–305. • Empirically established and validated: The framework is Boynton, A. C., & Zmud, R. W. (1984). An assessment based on a diverse set of companies and projects as of critical success factors. Sloan Management Review, 25(4), well as real-world data. It also takes a multi-level view 17–27. (no single-source bias), an approach that enabled us Conant, J. S., Mokwa, M. P., & Varadarajan, P. R. (1990). to develop a strong theoretical framework. Strategic types, distinctive marketing competencies and • Contingent: The framework captures different configu- organizational performance: A multiple measures-based rations of project management elements to account study. Strategic Management Journal, 11(5), 365–383. for specific business strategies (differentiation, cost Cooper, R. G., Edgett, S. J., & Kleinschmidt, E. J. leadership, and best-cost), and thus presents a contin- (1998a). Best practices for managing R&D portfolios: gency approach based on the differences. Research. Technology Management, 20–33. Cooper, R. G., Edgett, S. J., & Kleinschmidt, E. J. Research Limitation (1998b). Portfolio management for new products. Reading, MA: Although Eisenhardt (1989) argued that four to ten cases Perseus Books. provide a sufficient range of measure and for analytic gener- Dubin, R. (1978). Theory building. New York: The Free Press. alizations, one major limitation in our study is the relative- Eisenhardt, K. (1989). Building theories from case ly small number of cases that we used to develop the study research. Academy of Management Review, 14(4), framework (eight cases). This study may also suffer from a 532–550. bias of company management views. However, we were able Englund, R. L., & Graham, R. J. (1999). From experi- to minimize any such bias by using multiple data sources ence: Linking projects to strategy. The Journal of Product (review of related documents received from companies, the Innovation Management, 16(1), 52–64. existing literature among others) and validating the findings Hamel, G., & Prahalad, C. K. (1989). Strategic intent. with a panel of experts. Harvard Business Review, 67(2), 92–101. Harris, J. R., & McKay, J. C. (1996). Optimizing product Future Study development through pipeline management. In D. R. Rosenau The research findings and limitations suggest that the align- (Ed.), The PDMAA handbook of new product development (pp. ment measurement methodology deserves an empirical 63–76). New York: Wiley. study. If such a study uses a comprehensive approach, Harrison, F. L. (1992). Advanced project management: A researchers could standardize the measurement and create a structured approach (3rd ed.). New York: Halsted Press. framework for comparative studies of aligning the various Harrison, J. S., & St. John, C. H. (1998). Strategic manage- project and business strategy types. This would also enable ment of organizations and stakeholders (2nd ed.). Cincinnati, OH: researchers to work toward determining the degree of align- South-Western College Publishing. ment required to assure project and business success in rela- Hartman, F. (2000). Don’t park your brain outside: A practition to different circumstances. Researchers should apply such cal guide to improving shareholder value with SMART management. a contingency approach in subsequent studies. What is also Newtown Square, PA: Project Management Institute.

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 109 Hill, C. W. L. (1988). Differentiation versus low cost or dif- Reitsperger, W. D. (1993). Product quality and cost leader- ferentiation and low cost: A contingency framework. Academy of ship: Compatible strategies? Management International Review, Management Review, 13(3), 401–412. 33(1), 7–22. Jamieson, A., & Morris, P. W. G. (2004). Moving from cor- Shenhar, A. (1999). Strategic project management: The new porate strategy to project strategy. In P. W. G. Morris and J. K. framework. In Portland International Conference on Management of Pinto (Eds.), The Wiley guide to managing projects (pp. 177–205). Engineering and Technology (PICMET) Proceedings (pp. 382–286), Hoboken, NJ: John Wiley & Sons, Inc. Portland, OR. Kald, M., Nilsson, F., & Rapp, B. (2000). On strategy and Shenhar, A. J. (2001). One size does not fit all projects: management control: The importance of classifying business. Exploring classical contingency domains. Journal of the Institute British Journal of Management, 11, 197–212. for Operations Research and the Management Science, 47(3), Kathuria, R., & Davis, E. B. (2001). Quality and work force 394–414. management practices: The managerial performance implica- Srivannaboon, S., & Milosevic, D. Z. (2004). The process of tion. Production and Operations Management, 10(4), 460–477. translating business strategy in project actions. In D. P. Slevin, D. Kim, L., & Lim, Y. (1988). Environment, generic strategies, I. Cleland, & J. K. Pinto (Eds.), Innovations: Project management and performance in a rapidly developing country: A taxonomic research 2004 (pp. 175–192). Newtown Square, PA: Project approach. Academy of Management Journal, 31(4), 802–827. Management Institute. Miles, R. E., & Snow, C. C. (1978). Organizational strategy, Strauss, A., & Corbin, J. (1998). Basics of qualitative research: structure and process. New York: West. Techniques and procedures for developing grounded theory (2nd ed.). Miller, A., & Dess, G. G. (1993). Assessing Porter’s model Thousand Oaks, CA: Sage Publications. (1980) in terms of its generalizability, accuracy and simplicity. Thompson, A. A., & Strickland, A. J. I. (1995). Crafting and Journal of Management Studies, 30(4), 553–585. implementing strategy. Chicago: Irwin. Miller, D., & Friesen, P. (1986). Porter’s (1980) generic Treacy, M., & Wiersema, F. (1995). The discipline of market strategies and performance: An empirical examination with leaders. Reading, MA: Addison Wesley. American data. Organizational Studies, 7(1), 37–55. Tse, E. C., & Olsen, M. D. (1999). Strategic management. In Milosevic, D. Z. (2003). Project management toolbox: Tools B. Brotherton (Ed.), The handbook of contemporary management and techniques for the practicing project manager. Hoboken, NJ: research (pp. 351–373). New York: John Wiley & Sons. John Wiley & Sons. Turner, J. R., & Simister, S. (2000). The Gower handbook of Mintzberg, H. (1994). The rise and fall of strategic planning. project management (3rd ed.). Aldershot, UK: Gower. New York: The Free Press. Veliyath, R. (2000). Firm capabilities, business strategies, Papke-Shields, K. E., & Malhotra, M. K. (2001). Assessing customer preferences, and hypercompetitive arenas: The sustain- the impact of the manufacturing executive’s role on business ability of competitive advantages with implications for firm com- performance through strategic alignment. Journal of Operations petitiveness. Competitiveness Review, 10(1), 56–83. Management, 19(1), 5–22. Wheelwright, S. C., & Clark, K. B. (1992). Revolutionizing prod- Pinto, J. K., & Covin, J. G. (1989). Critical factors in project uct development. New York: The Free Press. implementation: A comparison of construction and R&D proj- White, R. E. (1986). Generic business strategies, organiza- ects. Technovation, 9(1), 49–62. tional context and performance: An empirical investigation. Phillips, L. W., Chang, D. R., & Buzzell, R. D. (1983). Product Strategic Management Journal, 7(3), 217–231. quality, cost position, and business performance: A test of some key Youndt, M. A., Snell, S. A., Dean, J. W., & Lepak, D. P. (1996). hypotheses. Journal of Marketing, 47(2), 26–44. Human resource management, manufacturing strategy, and firm Porter, M. E. (1980). Competitive strategy. New York: Free Press. performance. Academy of Management Journal, 39(4), 836–866.

DRAGAN Z. MILOSEVIC, PMP, is an associate professor of engineering and technology management in the Department of Engineering and Technology Management at Portland State University, where he has developed practical tools and innovative approaches to the traditional and current challenges of project management, and published numerous papers in management journals. He is a leading authority on project and program management, and has more than 20 years of experience in international work as a project manager and distilled some of his theoretical ideas and practical knowledge into an award-winning book, Project Management Toolbox, published in 2003 by Wiley. Milosevic earned his credentials as a project manager in the private sector, managing large projects around the world. As a consultant with Rapidinnovation, LLC, an executive consulting company, he helps leading companies streamline their project and program management models to ensure profitability. He also conducted project management seminars for the Project Management Institute, where he has earned their project management professional designation of PMP. He has worked in this field at a wide range of companies including Intel, Hewlett- Packard, Armstrong World Industries, Boeing, Daimler Chrysler, Mentor Graphics, Tektronix, Inc., InFocus, Credence Systems, Flextronics, WelchAllyn, Tyco, and U.S. Sprint.

SABIN SRIVANNABOON received a PhD in systems science in engineering management from Portland State University (PSU). He has an MS in engineering management from PSU, and BE in environmental engineering from Chulalongkorn University (Thailand). He currently works as an NPI program manager for a hi-tech company in Hillsboro, OR, USA. As an academician, Sabin presented his research at well-recognized international technology and project management conferences, where many of his articles were selected to be published as chapters in books. In addition, Sabin has published articles in International Journal of Project Management, Engineering Management Journal, and Project Management Journal. Sabin was listed in the 2005-2006 “Honors Edition” of the United Who’s Who Registry of Executives & Professionals as a scholar/ professional who has demonstrated outstanding leadership and achievement in his profession. He also won an “Outstanding Graduate Student Award” from Maseeh College of Engineering and Computer Science, PSU.

110 AUGUST 2006 PROJECT MANAGEMENT JOURNAL NEW POSSIBILITIES FOR PROJECT MANAGEMENT THEORY: A CRITICAL ENGAGEMENT

SVETLANA CICMIL, Bristol Business School, University of the West of England Bristol, U.K.

DAMIAN HODGSON, Manchester Business School, University of Manchester, U.K.

Introduction ABSTRACT oncepts developed by the academic community… must be recovered “ from operational and textbook definitions and reconnected to ways of This paper provides avenues for a broader C seeing and thinking about the world. In the dialectics of the situation engagement with the conceptual considerations of projects and project management and the talk of individuals with different perspectives, the emergence of new ways with the aim of creating new possibilities of talking becomes possible” (Alvesson & Deetz, 2000, p. 146). for thinking about, researching, and devel- Several prominent authors (Koskela & Howell, 2002; Maylor, 2001; Morris, oping our understanding of the field as 2004; Morris, Patel, & Wearne, 2000; Winch, 1996) have raised the need to intro- practiced. Attention is drawn to the legacy of conventional but deeply rooted main- duce alternative theoretical approaches to the study of projects, and to identify the stream approaches to studying projects implications that they may have for how we organise and manage projects. The pur- and project management, and implications pose of this paper is to address this need, by identifying space outside of the tight- of the specific underpinning intellectual ly-defined and densely populated conceptual landscape of mainstream project tradition for recommendations proposed to organisational members as best practice management where other perspectives, other concerns, and other agenda may be project management. The identified con- articulated and explored1. Extant project management literature, we would argue, cerns and limitations are discussed in the tends to rely upon the language of design, regularity and control to propose mod- context of project management evolution where taken-for-granted advantages of els and prescriptions as a route to increasing the ability of humans to control com- project management as a disciplined effec- plex worlds (Stacey, 2001; Wood, 2002), to the exclusion of other approaches or tive methodology and its popularity are ways of reasoning. As a whole, research into projects and project management reexamined. The paper sheds light on a remains heavily reliant on a functionalist, instrumental view of projects and organ- variety of voices from both scholarly and practitioner communities that have isations, where the function of project management is taken to be the accomplish- attempted to respond to this paradox and ment of some finite piece of work in a specified period of time, within a certain move the field forward. Taking issue with budget, and to agreed specifications. Most textbooks and professional associations conventional labels of project success or for project management promote this normative view of the field as practiced, failure, and drawing attention to alternative theoretical and methodological proposi- which can be summarised as the application of knowledge, skills, tools, and tech- tions, the argument turns toward critical niques to project activities to meet project requirements. Governed by the tradition management studies, outlining the impli- of “natural sciences” (e.g., systems theory), the project management body of knowl- cations of this intellectual tradition for stud- edge emphasises the role of project actors and managers as “implementers” nar- ies of projects, project management, project performance, and individual skills rowing their role to the issues of control (time and cost) and content (planned and competencies to cope with social scope of work), marginalising their wider potential role as competent social and arrangements labelled “projects.” political actors in complex project-labelled arrangements. Dissemination of “best Keywords: project management theory; practice” carries a message about the possibility of the progressive rationalisation of project management methodology; action and a belief in the progressive and cumulative character of knowledge. This research and development typically assumes rationality, universality, objectivity, and value-free decision-mak- ©2006 by the Project Management Institute ing, and the possibility of generating law-like predictions in knowledge. Vol. 37, No. 3, 111-122, ISSN 8756-9728/03

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 111 The limitations and challenges to this view of projects As a tentative starting point, therefore, we would pose some are, however, widely recognised across the field, and increas- fundamental questions that might guide our reflection on how ingly within the project management community itself. projects are conceived and how they could be conceived: Project management has attracted significant attention from • Is there a universal explanation of what projects are an increasing number of researchers and practitioners across and how projects evolve? management disciplines, coincident with the increased • What is the meaning behind the concepts in use, that “adoption” of project-based work across industrial sectors is, the terms such as “project”, “project management,” (Cicmil, 2001; Hodgson, 2002; Kreiner, 1995; Packendorff, and “project success”? 1995). At the same time, the foundations and practical appli- • What are the implications of the “mainstream” defi- cation of this managerial technology, embodying the scien- nitions of “project’”and “project management” for the tific achievements of operational research in work scheduling nature of knowledge and the intellectual foundations and control under specific constraints of time, cost, and a of studies of project-based organising, work, and unique outcome, have been seriously questioned by both the management? academic and practitioner communities. Several important • What are the consequences of project organising as writers in this field maintain that little radical examination of currently prescribed, both for project managers and the intellectual foundation of project management has been project workers? done within this stream of research, arguably since the 1960s • What alternative perspectives upon projects exist (Koskela & Howell, 2002; Morris, 1997). In the same vein, beyond the mainstream? writers such as Frame (1994, 1995, 1999), Morris (1997), • Whose interests are being served by the reproduction and Maylor (1999, 2001), among others, have called for a of the status quo in the field? reexamination of the dominant doctrines in project management for their failure to deliver on their promises. To understand why we have highlighted this sort of con- Nonetheless, there are limitations to this self-critique; the cern (and, equally importantly, why we feel such concerns tendency in the field is still to start from the assumption that are not routinely considered in the vast “mainstream” liter- the basic framework of project management is compelling ature on projects), we will need to locate our discussion and essentially sound. Efforts have therefore been directed within a reexamination of the evolution of project manage- instead towards searching for improvements in traditional ment. In doing so, we aim to underline why projects merit models and skills (see, for example, Maylor, 2003; Meredith such serious attention, and to account for their rising popu- & Mantel, 2003; Young 1999, 2003) towards a model that larity and importance in contemporary organisations. better represents the “true” nature of projects, or for a method of project management based on “critical success Project Management in Perspective factors” (Belassi & Tukel, 1996; Belout, 1998; Boddy & Project management emerged as a social practice in the Paton, 2004; Kharbanda & Pinto, 1996; Stallworthy & post-World War II development of technology and infra- Kharbanda, 1985), with the assumption that such an ideal structure. Although for many writers, project management model can objectively exist in the world of practice. There is has a much longer ancestry, traceable back to prehistoric little evidence yet that the resulting torrent of competing times, we would strongly oppose this ahistorical perspective streams of thought, methods of inquiry, and best practice (Hodgson & Cicmil, 2006b), which affords a spurious pedi- claims and propositions has creatively contributed either to gree to techniques, models, and procedures that have exist- constructive debate in the field or to the resolution of diffi- ed in something close to their current incarnation for culties encountered in practice. certainly less than a century. The emergence of project man- To address this situation, we intend through this paper agement is described in some detail by Morris (1997) and to create an opportunity to stand back and problematise Engwall (1995), highlighting its development in practice that which seems known and accepted about projects. through a number of major projects that can be traced back Taking this concern as its point of departure, our aim is to to the Manhattan project in the 1940s. Although the U.S. oil open up new trajectories within the research agenda in the and chemicals industry played a major role in this period, field of studies relevant to projects, project performance and the majority of the groundwork was done in U.S. defence project management (Hodgson & Cicmil, 2006a). The start- and aeronautics in the 1950s, including widespread use in ing objective is to critically evaluate the intellectual founda- the Apollo space programs (Harrison, 1981). As is evident tions of project management as a field of study and a from contemporary writings (Gaddis, 1959), the Cold War practicing discipline, to expose and understand the key acted as a significant driver on project management devel- obstacles to innovative research and the creation of knowl- opment in the U.S. throughout this period. The intellectual edge communicable and relevant to practitioners, and to activity in developing the field until the 1960s was based broaden the research agenda by encouraging a more critical almost exclusively on quantitative techniques within opera- approach in this area of organisational life. In particular, the tional research (OR). During the 1960s and 1970s, the pre- paper will explore the potential of critical research in dominantly technicist approach was criticised and the enhancing the intellectual basis of the project management theoretical foundations of the field expanded (Packendorff, subject area. 1995; Winch, 1996) to encompass traces of organisational

112 AUGUST 2006 PROJECT MANAGEMENT JOURNAL research and theories largely concerned with project organi- organization that enable concentration of flexible, sation structures (i.e., the matrix form), project leadership, autonomous, and knowledgeable individuals in temporary the role of human resource management in facilitating proj- project teams, for the focused accomplishment of goals effi- ect work and advice on project team building. In the 1980s ciently, timely, and effectively, for customer satisfaction and and 1990s, there was a revival of the OR-based project man- company benefits. Despite the various streams of praise and agement research driven by the developments of computer- criticism in the last 50 years, then, project management and based technology, which resulted in the creation and projects have now been accepted by many both within and promotion of sophisticated expert systems for project plan- outside the field as natural, self-evident, and indispensable. ning, control and risk analysis, and an increased use of ter- The promotion of projects and project management minology such as project information systems, project continues to expand as knowledge-intensive firms increas- communication networks, etc. This was in no small measure ingly based on project models have been acclaimed by many due to the awakening of public sector clients, including gov- as the organisation of the future (Frame, 1999; Weick, ernment agencies, in their search for robust management 1995). Thus, Frame, for example, claims confidently that the models and procedures to minimise disasters of budget and underlying reason for the projects becoming the central time overruns and questionable quality associated with the focus of management activity in many organisations can be project work and outcomes delivered by contractors. A vari- stated “in a single word: competition” (1999, p. 4, italics ety of project control methodologies (for example, the original). The literature since the 90s has drawn attention to PRINCE family) and risk management schemes have been the centrality of project-based organizing and project work- developed against such a background. Despite the increased ing in the processes of information sharing and knowledge sophistication of these models for project planning and management in organizations (Davenport & Prusak, 1998; monitoring, researchers found that only the most basic ones De Fillipi, 2001; Hansen, Nohria, & Tierney, 1999; Silver, are actually used by practitioners and that they are not 2000Wiig, 1997). Cleland (1997) pointed out that, as proj- always used as intended (Packendorff, 1995). ect teams evaluate new technologies and resources, they The 1990s saw an expansion of the project management gain insights into the need for making changes. Projects sup- field of study from its engineering heartlands into what posedly provide, according to Cleland, a central point where became widely accepted as a “multidisciplinary subject,” sig- new knowledge, skills, and attitudes can be developed. The nificantly engaging business and management researchers received wisdom of this kind has resulted in a widespread and educators (Winch, 1996). This coincided with the pro- adoption of “the project” in contemporary organizations as motion and acceptance of project-based work, organising, the focal unit of their operations. Not only are projects con- and management across industries and sectors, as a power- sidered suitable ways to control endeavours in a turbulent ful and universal organisational response to the challenges environment (Ekstedt, Lundin, Soderholm, & Wirdenius, of managing in a complex world. As Clarke (1999, p. 139) 1999), but also more importantly, they are regarded as the stated: “In a world where change is becoming increasingly appropriate way to stimulate a learning environment and important, tools such as project management, if used prop- enhance creativity so as to deliver complex products erly, can provide a useful way for organisations to manage (Hobday, 2000). Despite the inherent contradiction that change effectively.” It is usually based on the introduc- between these two arguments for project-based organising tion of a set of procedures, or on a new model of adminis- (Tjaeder & Thomas, 2000), it is precisely upon this ambi- tration with the strategic aim to enhance competitiveness tious promise to deliver both “controllability and adven- through a more effective intra-organisational integration ture” (Sahlin-Anderson & Söderholm, 2002) that the and optimal utilisation of scarce resources (Cleland, 1997). attraction of organisational “projectification” is founded. The contemporary surge in interest in project manage- In certain academic circles, the expanding influence of ment is typically explained by reference to the increasing “project-based work” has been referred to as the projectifica- recognition of “the project” as a versatile, flexible, and pre- tion of society (Jessen, 2002; Lundin & Söderholm, 1998; dictable form of work organisation. Its image as a universal Midler, 1995; Sydow & Staber, 2002). In essence, this notion solution to organisational problems has been established attempts to capture the growing colonisation of all quarters on the promotion of specific techniques for planning, mon- of life by project-related principles, rules, techniques, and itoring, and control, tried and tested in the operations of tra- procedures, aspiring to form a new “iron cage” of project ditionally project-oriented industries such as defence, rationality (Hodgson & Cicmil, 2003). As more and more aerospace, and construction (see, for example, Frame, 1999; organisational members are consequently being redefined Maylor, 2001; Young, 1999). Projects and project teams have as project workers and project managers across industrial sec- been hailed in both practitioner and academic discourses as tors, both scholarly and practitioner communities are unique economic and social processes on which the emerg- reflecting upon the implications of this shift for employees ing “knowledge economy” heavily relies (Briner & Hastings, and organisations (Hodgson, 2002; Packendorff, 1995), 1994; Clarke 1999; Cleland, 1997; Cleland & Ireland, 2002; particularly in terms of the impact on workplace identity, Frame, 1994, 1995; Meredith & Mantel, 2003; Young, reshaped intersubjective interaction, and increased control 2003). They are promoted as universally applicable tem- over the individual through ideologies of efficiency and per- plates for integrating, by design, diverse functions of an formativity (Fournier & Grey, 2000). These mechanisms are

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 113 actualised in a number of project-related contemporary ten- A glance at the content of recent public reports and dencies including the use of information technology (IT) in those recorded in previous studies (e.g., Atkinson, 1999; business process restructuring, the promotion of self-man- Ewusi-Mensah & Przasnyski, 1997; Flyvbjerg, Bruzelius, & aging teams, the ideology of the “knowledge society” and Rothengatter, 2003; Flyvbjerg, Holm, & Buhl, 2002; Morris the “knowledge worker,” and the emergence of the project- & Hugh, 1987; Standish Group, 1995; Williams, 1999; based organisation. The resulting drive towards the profes- Winch, 1996) provides an insight into frequent cost over- sionalisation of the project management discipline has been runs, delays, and underperformance in terms of quality and accompanied by the struggle and tensions involved in con- user satisfaction, which seem to have become the rule and ceptualising, promoting, and agreeing on the universally the reality of contemporary projects. In 1995, for instance, it acceptable document that should outline the formal body was estimated by the Standish Group that American compa- of project management knowledge. This struggle reflects and nies and government agencies spent US$81-billion on can- encapsulates the competition between the nationally- celled IT projects (Ewusi-Mensah & Przasnyski, 1997). In embedded professional associations of project manage- addition, the same source reports that, in total, 31% of IS/IT ment, with distinct bodies of knowledge proposed by the projects were deemed complete failures; 53% were late, over (U.S.-based) Project Management Institute, but also the budget, and did not meet expectations; only 9% of IT proj- (U.K.-based) Association for Project Management, the ects were delivered on time or within budget; and a mere Japanese Engineering Advancement Association (ENAA), 16% were considered successful. The average time overrun the International Project Management Association has been identified as being 222% of the original estimate. (European in origin), in addition to the British Standards The question often raised in public about this issue is, gen- Institute (BSI) Guide to Project Management (BS6079) and erally, how the IT/IS project risk (both financial and service the numerous corporate models such as Ericsson’s “PROPs” risk) is shared and transferred among the public sector and model (Linde & Linderoth, 2006). Despite the ongoing private sector participants. debates, however, project management is the focus of In January 2000, The Financial Times reported, for exam- unprecedented interest in the first decade of the new mil- ple, on the “fiascos” of the major government IT projects in lennium. This interest in many senses flies in the face of the the U.K. “stemming from basic project errors” that “high- question marks that remain over the effectiveness of project lighted the need for greater professionalism in project man- management, the theoretical underpinnings of project management.… The government’s track record in project agement theory, and the transferability of project manage- management has been, to say the least, poor” (Whitehall, ment into novel industrial sectors and organisations. 2000). Here, the blame was attributed to a lack of specialist project management knowledge among some civil servants Project Management in Question and ministers, and to different approval systems, which In the closing decade of the 20th century, project manage- have, according to some observers, resulted in unrealistic ment was challenged more seriously than in any previous project deadlines. A growing body of evidence shows that period. Despite the levels of research founded on the pre- similar observations and conclusions have been made in sumptions of instrumental rationality in decision-making relation to IT/IS in other sectors and types of organisations. and control, it is increasingly apparent that accepting and It is not only the poor performance of IS/IT projects that has applying such orthodoxy does not eliminate project failures, come under public scrutiny. Bowen, Clark, Holiday, and nor does it guarantee project success (Williams, 2004). Wheelwright (1994) reported that nearly 30% of product Although the project management body of thought has development projects never live up to business objectives. been substantially modified over the last decade, the core According to Winch (1996), U.K. government-procured concerns continue to shape academic enquiry and practi- construction projects ranging from hospitals to roads, suffer tioners’ discourses about projects and project management. from, on average, 14% cost overrun and 11% time overrun. Contemporary studies of project performance continue to More recently, the £214m refit project of the Royal Opera indicate the disparity between the maturing body of project House in Covent Garden resulted in a cancelled opening management know-how and the effectiveness of its applica- performance, and the remaining shows being run at huge tion (Atkinson, 1999; Belassi & Tukel, 1996; Baker, Murphy, technical risks associated with the operation of newly & Fisher, 1983; Morris et al., 2000; Williams, 1995), as an installed but not properly tested and mastered backstage increasing visibility is being given to the claims about proj- equipment (Royal Opera House, 1999). The Jubilee line ect and project management failures, and about dissatisfac- extension project for the London Underground, for exam- tion with project performance and outcomes by affected ple, has been characterised as having been “…a long saga of stakeholders. Simultaneously, a growing body of literature, overshot deadlines and overspent budgets” (Winder, 1999, as well as a growing body of empirical evidence and the p. 8). In the United States, the belated opening of Denver voices of numerous practitioners, supports the view that the International Airport after four embarrassing postpone- very reason for using projects and project management as a ments, various scandals, and a final cost of US$5 billion methodology for organisational innovation and change is at against the budgeted US$1.5 billion, has been held up as yet the heart of project failures (Clarke, 1999; Maylor, 2001; another example of project failure (Dempsey, Goetz, & Thomas, 2000). Szyliowicz, 1997). The much-derided construction of the

114 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Scottish Parliament Building was described by the Fraser and relationship with other aspects of organisational and report in September 2004 as being two-and-a-half years managerial reality (Frame, 1999; Meredith & Mantel, 1995, behind schedule with costs running approximately 10 times 2003; Morris et al., 2000; Walta; 1995; Wideman, 1995; more than the original estimate of £40m. The conclusion among others). Despite the significant presence of project- drawn about the destiny of such projects as a rule, inherent based working and organising across industrial sectors and in their very nature, is simple: “These projects never go the problematic qualifications of project outcomes as suc- according to plan” (Royal Opera House, 1999). cess or failure, a number of authors note that the develop- In light of this, it is unsurprising that governments are ment of project management knowledge remains unstable taking a greater interest than ever before in project manage- and fragmented. As a consequence, the dream of establish- ment, in an attempt to address this apparently perennial ing project management as an exemplary field of manage- failing of project management techniques. As previously ment science is becoming increasingly remote. Questions noted, governments, and, in particular, the U.S. government, have been raised about the underlying belief system that has been closely involved in the development of project exhibits a strong bias towards functionalist/unitarist tradi- management models and techniques for over half a century. tion, reductionism, operational research, and “how-to-do” The U.K. Office of Government Commerce (OGC) within prescriptive forms of intellectual output (Buchanan & the British Treasury, for example, has developed and pro- Badham, 1999; Kreiner, 1995; Packendorff, 1995). It was moted the well-established PRINCE and PRINCE2 models, mainly in the 1990s that critical analysis of social and polit- and is currently taking the lead in setting up Centres of ical power associated with projects as organisational and Excellence for Project and Programme Management social arrangements, and project management as a practice throughout the U.K. Similar initiatives by government agen- and as a social grouping emerged in an explicit form cies in North America and elsewhere indicate the serious- (Buchanan and Badham, 1999; Buchanan & Boddy, 1992; ness with which project management models and practices Kreiner, 1995; Lundin & Hartman, 2000; Lundin & Midler, are now considered in the public sector. Meanwhile, the 1998; Packendorff, 1995). ongoing professionalisation of the field of project manage- Nonetheless, the response to this crisis has so far been a ment, and the increased influence of professional associa- yet-greater emphasis on technicist solutions, quantitative tions through accreditation of training and credentialism, methodologies, positivist methodologies and a stronger draws significant support and gains moral legitimacy from reliance on instrumental rationality. In one attempt to move this perceived role in protecting public interests and ensur- the field forward, Atkinson (1999) asserted that it has ing the effective use of public funds. become an impossible, and, most likely, non-“value-adding” endeavour to define project management in terms of the tra- Diagnoses and Prescriptions ditional “iron triangle” principles, emphasising the achieve- Although the existence of a crisis of some kind in the field ment of time, cost, and quality objectives as the major of project management is recognised in many (although not justification for the role of project management. According all) quarters, the diagnoses in the field are unsurprisingly to Atkinson, the attention should be refocused from these varied. For many established project management writers, efficiency measurements, which are being questioned as the failings of project management are to be expected in a appropriate measures of project success (see also Baldry, maturing field. As techniques are further honed, and mod- 1998; Belassi & Tukel, 1996; Chapman, 1998; Maylor, 2001). els are perfected through longitudinal and cross-sectoral The issue of ambiguity associated with qualifying a proj- research, it is assumed that the field will one day settle upon ect as success or failure has recently attracted scholarly atten- a reliable and basically effective model and array of tech- tion (e.g., Boddy & Paton, 2004; Buchanan & Badham, niques. Others see the problem as far more deeply rooted 1999; Clarke, 1999). The debate focuses on a more strategic in the fundamental principles upon which the field of level of decision-making, in which project failure appears to project management has been established. In this section, be “strategic” rather than linked to technical problems, and we will look at each of these accounts in turn, considering is seen as a result of political processes of resistance in first the attempts by the mainstream of project manage- organisations. Accounts pointing to the evidence of mount- ment to confront the very real failings in the discipline, ing dissatisfaction and lack of support imply a range of before turning to a more critical diagnosis of the current social and behavioural factors behind project failures. Other state of project management. authors suggest the need for a wider picture of what goes on It is not our claim, then, that project management as a in social construction of projects and project management disciplinary area is unaware of, or unconcerned by, the lim- by focusing on who is included in, and who is excluded itations and continued failings of project management from, the decision-making process, analysing what deter- models and methods. There is a long-standing debate on the mines the position, agendas and power of different partici- international scene about the formulation of the formal, pants with respect to issues, and how these different agendas professional project management body of knowledge, in are combined and resolved in the process by which the deci- which important questions are posed by the proponents of sions are arrived at (Flyvbjerg et al., 2003). In the context of project management about the boundaries of the project IT/IS project failures, Taggert and Silbey (1986) cynically management subject area, its purpose, practical application, propose a political-development cycle of projects: wild

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 115 enthusiasm, disillusionment, total confusion, search for the process, and narrative. The first two essentially reflect the guilty, punishment of the innocent, and promotion of non- idea of Sauer (1999) who differentiates between a factor- participants, in contrast to the conventional rational project based approach and a process approach to project failure, life cycle (PLC) model, which neatly unfolds as a succession which is largely present in the extant project management of stages: conception and feasibility study, requirement body of thought. The third (narrative—Fincham, 2002) is analysis and specification, design and development, imple- related to a much less frequent approach to project per- mentation/execution, and project termination. formance research, but is the one that is of most interest to On the basis of his research into IT projects, Fincham this paper. (2002) argued that project failure can be interpreted in a The exploration of the performance characteristics of wide range of ways, joining the writers who focus on politi- public initiatives, such as large-scale engineering projects, cal discourses, language and interpretation related to organ- has expanded these performance measures to include a dis- isational reporting on project performance. Arguing that all cussion of the role of institutions, risks, and governance in projects exhibit, to a smaller or larger extent, a dimension of project success (Flyvbjerg et al., 2003; Miller & Lessard, organisational innovation and change, Fincham suggested 2000). For others, it is the paradoxical feedback that hinders that the attributions of “success” and “failure” can also be the effective adoption of project-based working and organ- explored as narratives that are involved in many forms of ising as a structural innovation in complex business envi- change and innovation in organisations. “Through a kind of ronments. Whereas project management has been social labelling events are formulated into evolving ‘stories’ mobilised as a blueprint for structuring and coordinating that evoke either status or stigma and play a powerful role in organisational change, according to Clarke (1999): “People ordering behaviour” (Fincham, 2002, p. 1). From that per- often do not see project management as something to help spective, the management of projects and the change associ- them but rather something which is mandatory, serving lit- ated with its initiation and outcomes is equated with the tle useful purpose” (Clarke, 1999, p. 144). Clarke identified “management of meaning” (Huczynski & Buchanan, 2001), the following as problematic in the application of project or with symbolic attempts to legitimise project proposals management as a vehicle of change; the rigid “standardisa- and particular definitions of problems and solutions in the tion” of project management as the mode of change man- face of competing ideas. Similarly, Fincham (2002) agement that often causes cultural clashes; project observed that “[t]he gloss of success is often critical for proj- management, or “managing by projects” or becoming a ects that may involve large expenditures and much uncer- “project-based” organisation is often regarded as another tainty” (p. 1), with expectations to justify huge, risky control mechanism, a “corporate reporting” tool; the inade- investments. Such a frame of reference emphasises the need quate formal completion of change projects; project over- to understand organisational processes of power, politics, load syndrome; individual resistance to imposed procedures structure, and their interplay with the nature of interaction and practice, and a lack of confidence and motivation. It among individuals and interest groups involved in develop- becomes obvious that, frequently, the very principles of ment and implementation of a project. effective, structured project management methodology are Table 1 summarises different approaches to under- simultaneously its major causes of failure. standing project failure by distinguishing three perspectives Another influential attempt to address the malaise and linking them to a wider domain of the project manage- comes from what is known as the “Scandinavian School” of ment process. This table has been adapted from Fincham project studies (Ekstedt et al., 1999; Lundin & Hartmann, (2002), who termed the three approaches as rationalist, 2000; Lundin & Midler, 1998; Lundin & Söderholm, 1995;

Perspective Form of Organisational Methodological Success and Failure Behavior and Action Focus Seen As Rational/normative Organisational goals; Simple cause and effect Objective and polarised managerial and states organizational structures surrounding the project

Processual Organisational and socio- Socio-technical interaction Outcomes of organizational political processes; processes projects as form of a decision outcome

Narrative Organisational and socio- Interpretation and sense- Social constructs; paradigms political processes; symbolic making; rhetoric and action; themes persuasion; critical/ hermeneutics

Table 1: Perspectives on project success and failure (adapted from Fincham, 2002, p. 3)

116 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Sahlin-Andersson & Söderholm, 2002; Söderlund, 2004). In Our proposal in this paper is to draw on the insights the 1990s, Packendorff succinctly summarised the concern offered by writers and academics within the broad grouping shared by a group of Scandinavian scholars, arguing that of critical management studies to widen and deepen the the- contemporary propositions for the improvement of project oretical foundations of project management and to provide management knowledge and practice were ill-conceived, novel approaches to the entrenched challenges facing proj- reflecting fundamental misconceptions within the field. He ect management. We foresee that this would require a fun- identified three major deficiencies which are ingrained, damental reappraisal of many of the core tenets of project maintained, and reproduced across the research field management theory and techniques, a stance that may be through certain ontological, epistemological, and method- uncomfortable for many whose careers and indeed liveli- ological assumptions: (1) the assumed universality of proj- hoods are intimately connected to project management as it ect management theory, (2) the lack of empirical studies of stands. Nonetheless, we argue that without such a radical projects, and (3) the lack of alternative representations of reappraisal of the field, project management as a discipline “projects” (summarised in Table 2), will not surmount the practical and philosophical concerns Emerging from this school are a number of vital themes that it currently faces, and will struggle to exercise a signifi- that move beyond traditional understandings of projects cant influence on the ever-wider application of project and their management: the conceptualisation of projects as organising in the years to come. temporary organisations (Lundin & Söderholm, 1995); the recognition of the historically-embedded nature of projects Critical Perspectives on Projects (Engwall, 2003; Kreiner, 1995); and the shift in focus from Critical work on management and organisations has a single to multiple project management (Engwall & Jerbrant, broad, even eclectic base, and draws upon a wide range of 2003). Although this introduction of sociological perspec- social theories, philosophies, and ethical/moral positions, tives to the field of projects is clearly welcome—indeed, including Marxism, feminism, environmentalism, labour long overdue—the more conservative current work in this process theory, post-structuralism, post-colonialism and tradition remains strongly wedded to a functionalist view- critical realism. Work in this field draws its inspiration and point, focusing upon improving project performance theoretical frameworks from a wide range of writers, includ- through attention to social (i.e., human) factors. Equally, ing Weber, Braverman, Derrida, Latour, Bourdieu, within this perspective, the inclusion of power and power Baudrillard, Foucault, Habermas, Bhaskar, among others. In relations tends to be limited to the introduction of a form of what stands as the most coherent summary of the principles micropolitics, separated from the larger power differentials underpinning this diverse field, Fournier and Grey (2000) inherent in modern, capitalist society. Although there is set out three key tenets that critical work shares. They argue much to be drawn from the Scandinavian School, and much that critical research on management and organisations: that improves upon the narrow mechanistic instrumental- 1. Has a “Non-Performative Intent”—Starts out from the ism of traditional project management, we would argue that position that issues of morality, equality and ethics are the school remains too conservative in its ambitions, and at least as important as, and in many senses more does not take its argument to its logical conclusions. It does, important than, the traditional functionalist concerns however, open the space of project studies to more explicit- of effectiveness and efficiency of management. ly “critical” currents, and some of the strongest critical work 2. “Aims to Denaturalise Organisations & on projects to have emerged so far has its roots in the Management”—Challenges arguments that the cur- advances made by the Scandinavian School (see, for exam- rent way in which organisations, economies, and ple, Lindgren & Packendorff, 2001, 2006). societies are organised is somehow natural, normal

Common Assumptions Alternative Assumptions

Project management theory General theory for all kinds of projects, Middle-range theories on different generic concept collecting different sorts of projects, classified according theories applicable to projects under to different selection criteria one umbrella

Aim of research on projects Prescriptive, normative theory, Descriptive theory, grounded in grounded in ideal models of project empirical narrative studies on human planning and control. Research interaction in projects. Research undertaken as survey studies of large undertaken as comparative case samples of projects studies

Research metaphor for the project A tool, a means for achieving higher- A temporary organisation, an level ends aggregate of individuals temporarily enacting a common cause

Table 2: Common and alternative assumptions on project management (adapted from Packendorff, 1995, p. 326)

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 117 or inevitable, arguing instead that the status quo is a the development of critical project management studies is consequence of the prioritisation of the agenda of the inclusion of critical social theory into the research certain social groups, and benefits these groups at the process. Central to it is the need to explore how the rela- expense of others. tionships between individuals and collectivities are being 3. “Aims to Prevent Oppression/Exploitation”—The constituted and reproduced in the context of project man- overarching mission underpinning critical work is to agement, and how asymmetrical power relations create and highlight and oppose oppression and exploitation in sustain the social reality of projects. In this context, we argue organisations and societies. Typically, this refers to that the main issue for the project management research the exploitation of employees, of women, of ethnic community should not be what form of critical analysis is minorities, or of the environment. best suited for enhancing the intellectual basis of critical management studies. Drawing on critical theory and partic- Much of this work (for example, Alvesson & Willmott, ularly the contribution of Jurgen Habermas, Alvesson, and 1996; Reed, 1992; Thompson & McHugh, 2002) has since Willmott (1996) suggest that intellectual efforts should be the mid-1990s been collectively referred to as critical man- focused on encouraging inspiration from a variety of theo- agement studies(CMS), which has also lent its name to a ries and ideas, as a counterforce to technicist and instru- major international conference (CMS5, to be held in 2007), mental forms of rationality in project environments. From a as well as to a major interest group at the Academy of Habermasian perspective, it might be argued that the objec- Management. Such work takes issue with positivist episte- tive, abstract and universal body of knowledge claimed in a mology in the field of management, insofar as it perpetuates number of authoritative sources as proprietary to project the belief that managers face an objective reality that they can management fails to live up to the challenges of the embod- control by applying suitable methods for a rational assess- ied and power-laden realities of its operation. “Project man- ment of the problematic situation in order to come up with agement” as created by this school of thought, exhibits the the correct solution. The consequence of this is a prolifera- characteristics of what Alvesson and Willmott (1996) called tion of methods, tools, analytical techniques, and applied management as colonising power and management as distorted instruments with which management itself becomes identi- communication. From this perspective, the possibility of crit- fied. Management skills and knowledge are reduced to ical project management will depend on the extent to which value-neutral competence, ignoring the political aspect of a social theory about the nature of projects provides con- organisations, and ethical and moral issues, reinforcing the cerned actors with authentic insights into their position in belief that management can be conceptualised in a technical project environments, leading to their enlightenment, way by agreeing on terminology and meaning. Managers are changed attitudes and emancipatory action. seen as rational technicians, dealing with technical issues Another major influence on critical work with implica- that are resolvable through the application of superior tions for an understanding of project management is the knowledge of the planning and control techniques. wide and varied oeuvre of Michel Foucault, drawn upon by Alvesson and Deetz have commented on the problems with writers such as David Knights, Stewart Clegg, Barbara narrow, conventional approaches to studying the phenome- Townley, and Stanley Deetz, among others. In particular, a non of management and the need to adopt a much more key research theme is a focus on the consequences of those critical stance and varying theoretical lenses: techniques of observation, measurement, and performance There is considerable agreement that conventional, univer- control central to project management methodologies for sal statements of what management is about and what both the management and the self-management of workers managers do—planning, organizing, coordinating and within project settings. Work on project management in this controlling—do not tell us very much about organisation- tradition (Hodgson, 2002; Lindgren & Packendorff, 2003; al reality, which is often messy, ambiguous, fragmented Thomas, 2003) tends to criticise the implied calculability and political in character (Alvesson & Deetz, 2000, p. 60). and formality of project management methodology, as it embodies a strong functionalist commitment to ensuring, It is argued that conventional approaches to organisa- first and foremost, the effective control of workers. In many tional and management research have exposed managers ways, as previously noted, this control imperative in project and other employees involved in problem-solving and deci- management is traditionally based upon similar principles sion-making to an overwhelming amount and range of tech- to those underpinning scientific management: the fragmen- niques (empowerment, teamwork, flexibility), which can be tation of work and the maximisation of visibility and interpreted as “covert tools of manipulation and exploita- accountability. However, with isolated exceptions (Metcalfe, tion” (Huczynski & Buchanan, 2001, p. xxi). 1997), the fundamentals of project work appear to have More work has emerged that applies this critical posi- evaded the practical and moral critique levelled at other tion to project management, its nostrums and methods (see, Taylorist work forms. At the same time, the ongoing profes- for example, Bredillet, 2002, 2004; Buckle & Thomas, 2003; sionalisation of project management can also be interpreted Cicmil, 2003; Cicmil & Hodgson, 2005; Gill, 2002; in line with other professionalisation projects, as a mode of Hodgson, 2002; Hodgson & Cicmil, 2003; Metcalfe, 1997; control over expert labour (cf. Larson, 1977; Abbott, 1988), Packendorff, 1995). The most important requirement for implementing and enforcing a form of self-disciplinary con-

118 AUGUST 2006 PROJECT MANAGEMENT JOURNAL trol over project managers. A key challenge for critical work project management as social phenomena through critical from this perspective is thus to draw attention to the power studies? How might a consideration of lines of reasoning relations established and maintained by project manage- and practice other than those promoted by narrowly instrument technologies, and the consequences of these power mentalist project management thinking respond to the relations. Other critical work on projects includes analysis of identified crisis in the field? the gendered nature of project management models (Buckle The first and most important consequence of an & Thomas, 2003), examinations of the impact of projects engagement with critical work would be an increased sen- working on work-life balance (Lindgren & Packendorff, sitivity to the possibility of oppression and exploitation in 2003), investigations into the role of project management in project settings, an outcome which is especially likely perpetuating oligarchic elites in modern corporations given the pressurised environment of most projects, (Clegg & Courpasson, 2004), and analyses of project man- regardless of sector and scale. A second important aspect of agement as a form of bureaucratic control in post-bureau- critical approaches to project management is to reexamine cratic contexts (Hodgson, 2004). the currently dominant imperative of performativity in To advance such critical work, we would argue that the relation to how this shapes the development of the body scope for critical research into projects and project manage- of knowledge and best practice in the field (particularly ment, and in particular the emancipatory aspect of such related to “critical factors for project success”) and illumi- research needs to be expanded. This means a more radical nate the importance of considering other indicators of acknowledgment of voices from practitioners in project “project success” beyond time, cost, and quality perform- environments, such as Balck’s: ance, to encompass environment, health and safety, econ- Practitioners, in particular we as project managers, are omy, and ethics. And, third, critical project management well advised to rid ourselves of the constricting histori- research would engage directly with not merely project cal background of a mechanistic world image and managers but with practitioners at all levels of the project rationalism. Without question the best method to help hierarchy, as it is interested in specific local situations and us correct our way of traditional thinking is “on-the-job the lived experiences of various project actors, often with training”—that is, experiencing the real success and fail- the aim of initiating some transformative redefinition ures in dealing with our everyday business endeavours (Alvesson & Deetz, 2000) of actors’ own perception of self, (Balck, 1994, pp. 2–4). their voice, and their influence in shaping their own social roles and place. We would argue that taking this seriously means mov- Our intention in presenting these views is primarily to ing beyond the narrow instrumentalism, which bedevils, yet start a dialogue, to raise the importance of such issues and largely defines, the “iron triangle” approach to project man- concerns within the field of project management, and to agement. Our hope is that this paper may encourage move- draw attention to broader theoretical resources available to ment towards the creation of a vocabulary and a resource for conceptualise projects and their management. To this end, a critical engagement between practitioners and academics we have in the past organised workshops to act as forums beyond the confines of the existing language, concepts and for debates around these themes, of which the best papers assumptions of project management. have been published as a collection (Hodgson & Cicmil, 2006b). We will continue to run these workshops, in the Making Projects Critical: New Trajectories hope of extending the debate and learning from the views, At this point, it is opportune to return to the intentions experiences and insights of as broad a range as possible, behind this paper. As stated in the introduction, we aim to and we would welcome interventions, participation, and open new trajectories within the research agenda in the field even counter-critique from all quarters. We hope that, at the of studies relevant to projects, project performance and proj- least, this paper has succeeded in raising new concerns and ect management in order to address the persisting and pre- has sharpened interest among project management aca- vailing concerns articulated in literature and practice, which demics and practitioners to engage with these concerns, we attempted to briefly illuminate. Our immediate objec- which are central to the future of a sustainable and ethical tive, therefore, is to signpost possible research trajectories project management. towards a critical evaluation of the intellectual foundations of project management as a field of study and a practising Notes discipline and to broaden the research agenda by encourag- 1 In this sense, the paper continues the mission of two working a more critical approach in this area of organisational shops organised and held at Bristol Business School in the U.K. life. Explicitly we take up the challenge offered by Flyvbjerg in 2003 and 2004, with the explicit aim to “make projects crit- (2001, p. 166) to conduct research that “contributes to soci- ical.” These workshops brought together a diverse community ety’s capacity for value-rational deliberation and action”; in of researchers and practitioners from Europe, North America, essence, to make social science matter in the context of proj- and Australasia with a common interest in considering vital ect work. What, therefore, might it mean for both the schol- issues and values that are both ignored and obscured by “main- arly and the practitioner community, to encourage a stream” project management. Details can be found at different way of viewing and thinking about projects and http://www.uwe.ac.uk/bbs/research/research/mpc/index.shtml

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 119 References nel on design production within construction projects. Abbott, A. (1988). The system of professions: An essay on the International Journal of Project Management, 16(4), 235–247. division of expert labour. London: University of Chicago Press. Cicmil, S. (2001). Reconstructing the project manage- Alvesson, M., & Willmott, H. (1996). Making sense of ment knowledge system: A multiple-perspective agenda. management: A critical introduction. London: SAGE. Paper presented to Critical Management Studies Alvesson, M., & Deetz, S. (2000). Doing critical manage- Conference, UMIST, Manchester, UK, 11th-13th July 2001. ment research. London: Sage. Cicmil, S. (2003). Knowledge, interaction and project Association for Project Management (APM). (1995). work: The perspective of complex responsive processes of Body of knowledge, Version 2, January 1995 APM: London. relating. Paper presented at the 19th EGOS Colloquium, Atkinson, R. (1999). Project management: Cost, time subtheme “Project Organizations, Embeddedness and and quality, two best guesses and a phenomenon; It’s time Repositories of Knowledge,” Copenhagen, Denmark, 3–5 to accept other success criteria. International Journal of Project July 2003. Management, 17(6), 337–342. Cicmil, S., & Hodgson, D. E. (2005). Knowledge, action, Baker, B. N., Murphy, D. C., & Fisher, D. (1983). Factors and reflection in management education—The case of project affecting project success. In D. I. Cleland, & R. W. King management. BBS Teaching and Research Review, Issue 7, (Eds.), Project management handbook (pp. 669–685). New September 2005; ISSN 1468-4578. Retrieved October 10, York: Van Nostrand Reinhold. 2005, from http://www.uwe.ac.uk/bbs/trr/Is7-cont.htm Balck, H. (1994). Projects as elements of a new indus- Clarke, A. (1999). A practical use of key success factors to trial pattern: A division of project management. In D. I. improve the effectiveness of project management. International Cleland & R. Gareis (Eds.), Global project management hand- Journal of Project Management, 17(3), 139–145. book (pp. 2-1–2-11). New York: McGraw - Hill International Clegg, S. C., & Courpasson, D. (2004). Political hybrids: Editions. Toquevillean views on project organizations. Journal of Baldry, D. (1998). The evaluation of risk management Management Studies, 41(4), 525–547. in public sector capital projects. International Journal of Cleland, D. I. (1997) Field guide to project management. Project Management, 16(1), 35–41. New York: Van Nostrand Reinhold. Belassi, W., & Tukel, O. I. (1996). A new framework for Cleland, D. I., & Ireland, L. R. (2002). Project management: determining success/failure factors in projects. International Strategic design and implementation (4th ed.). New York: Journal of Project Management, 14(3), 141–151. McGraw-Hill International. Belout, A. (1998). Effects of HRM on project effective- Davenport, T. H., & Prusak, L. (1998). Working knowledge. ness and success: Toward a new conceptual framework. Boston: Harvard Business School Press. International Journal of Project Management, 16(1), 21–26. DeFillippi, R. (2001). Project based learning, reflective Boddy, D., & Paton, R. (2004). Responding to compet- practices and learning outcomes. Management Learning, 32(1), ing narratives: Lessons for project managers. International 5–10. Journal of Project Management, 22(3), 225–233. Dempsey, P. S., Goetz, A. R., & Szyliowicz, J. S. (1997). Bowen, H. K., Clark, K. B., Holloway, C. A., & Denver International Airport: Lessons learned. New York: Wheelwright, S. C. (1994). The perpetual enterprise machine. McGraw-Hill. Oxford: Oxford University Press. Ekstedt, E., Lundin, R. A., Söderholm, A., & Wirdenius, H. Bredillet, C. N. (2002). Beyond the positivist mirror: Towards (1999). Neo-industrial organising London: Routledge. a project management ‘gnosis’. Paper presented at the IRNOP Engwall, M. (1995). Jakten på det Effektiva Projektet. conference in Rotterdam, Netherlands, May 28–31, 2002. Stockholm: Nerenius & Santérus. Bredillet, C. (2004). Beyond the positivist mirror: Towards Engwall, M. (2003). No project is an island: Linking proj- a project management ‘gnosis’. Paper presented at the IRNOP ects to history and context. Research Policy, 32(5), 879–808. conference in Turku, Finland August 25–27, 2004. Engwall, M., & Jerbrant, A. (2003). The resource alloca- Briner, W., & Hastings, C. (1994). The role of projects in tion syndrome: The prime challenge of multi-project manage- the strategy process. In D. I. Cleland & R. Gareis (Eds.), ment. International Journal of Project Management, 21(6), Global project management handbook (pp. 15_3–15_24). New 403–409. York: McGraw-Hill International Editions. Ewusi-Mensah, K., & Przasnyski, Z. H. (1997). Critical Buchanan, D. & Badham, R. (1999). Power, politics, and issues in abandoned IS development projects. Communications organisational change: Winning the turf game. London: Sage. of the Association for Computing Machinery, 40(9), 74–80. Buchanan, D., & Boddy, D. (1992). The expertise of the Fincham, R. (2002). Narratives of success and failure in change agent. Hemel Hempstead, UK: Prentice Hall. systems development. British Journal of Management, 13(1), Buckle, P., & Thomas, J. (2003). Deconstructing project 1–14. management: A gender analysis of project management Flyvbjerg, B. (2001). Making social science matter. guidelines. International Journal of Project Management, 21(6), Cambridge, UK: Cambridge University Press. 433–441. Flyvbjerg, B., Bruzelius, N., & Rothengatter, W. (2003). Chapman, R. J. (1998). The role of system dynamics in Megaprojects and risk: An anatomy of ambition. understanding the impact of changes to key project person- Cambridge, UK: Cambridge University Press.

120 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Flyvbjerg, B., Holm, M. S., & Buhl, S. (2002). Management, 11(4), 335–346. Underestimating costs in public works projects—Error or Larson, M. (1977). The rise of professionalism: A sociolog- lie? Journal of American Planning Association, 68 (Summer), ical analysis. Berkeley, CA: University of California Press. 279–295. Linde, A., & Linderoth, H. (2006). An actor-network Fournier, V., & Grey, C. (2000). At the critical moment: theory perspective on IT projects. In D. Hodgson & S. Cicmil Conditions and prospects for critical management studies. (Eds.), Making Projects Critical. Basingstoke, UK: Palgrave. Human Relations, 53(1), 7–32. Lindgren, M., & Packendorff, J. (2001). What’s new in Frame J. D. (1994). The new project management. San new organizational forms? On the construction of gender in proj- Francisco: Jossey-Bass. ect-based work. Paper for the Gender, Work and Organization Frame, J. D. (1995). Managing projects in organisations. conference ‘Rethinking gender, work and organisation’, San Francisco: Jossey-Bass. Keele University, Keele, UK, June 27–29, 2001. Frame, J. D. (1999). Project management competence: Lindgren, M., & Packendorff, J. (2006). Projects and Building key skills for individuals, teams and organisations. prisons. In D. Hodgson & S. Cicmil (Eds.), Making Projects San Francisco: Jossey-Bass. Critical. Basingstoke, UK: Palgrave. Gaddis, P. O. (1959). The project manager. Harvard Lundin, R. A., & Söderholm, A. (1995). A theory of the Business Review, May-June, 89–97. temporary organization. Scandinavian Journal of Gill, R. (2002). Cool, creative and egalitarian? Management, 11(4), 437–455. Exploring gender in project-based new media work in Lundin, R. A., & Söderholm, A. (1998). Europe. Information, Communication and Society, 5(1), 70–89. Conceptualising a projectified society: Discussion of an eco- Hansen, M. T., Nohria, N., & Tierney, T. (1999). What’s institutional approach to a theory on temporary organiza- your strategy for managing knowledge? Harvard Business tions. In R. A. Lundin & C. Midler (Eds.), Projects as arenas for Review, March–April, 1999. renewal and learning processes (pp. 13–23). Boston: Kluwer Harrison, F. L. (1981). Advanced project management. Academic Publishers. Aldershot, UK: Gower. Marshall, N. (2001). Knowledge, identity, and difference in Hobday, M. (2000). The project-based organisation: An project organisation. Paper presented at the 17th EGOS ideal form for managing complex products and systems? Colloquium ‘The Odyssey of Organising’, Lyon, France, 5–7 Research Policy, 27, 871–893. July 2001. Hodgson, D. E. (2002). Disciplining the professional: Maylor, H. (1999). Project management (2nd ed.). London, The case of project management. Journal of Management UK: Pitman Publishing. Studies, 39(6), 803–821. Maylor, H. (2001). Beyond the Gantt chart—Project man- Hodgson, D. E. (2004). Project work: The legacy of agement moving on. European Management Journal, 19(1), bureaucratic control in the post-bureaucratic organisation. 92–100. Organization, 11(1), 81–100. Maylor, H. (2003). Project Management (3rd ed.). London, Hodgson, D. E., & Cicmil, S. (2003, July). Setting the UK: Pitman Publishing. standards: The construction of ‘the project’ as an organizational Meredith, J. R., & Mantel, S. L. (1995). Project manage- object. Paper presented to the 3rd Critical Management ment—A managerial approach (3rd ed.). New York: John Studies Conference, Lancaster University, Lancaster, UK. Wiley & Sons. Hodgson, D. E., & Cicmil, S. (2006a). Making projects Meredith, J. R., & Mantel, S. L. (2003). Project manage- critical. London: Palgrave. ment—A managerial approach (5th ed.). New York: John Hodgson, D. E., & Cicmil, S. (2006b). Are projects real?: Wiley & Sons. The PMBoK and the legitimation of project management Metcalfe, B. (1997). Project management system design: A knowledge. In D. E. Hodgson & S. Cicmil (Eds.), Making social and organisational analysis. International Journal of projects critical. London: Palgrave. Production Economics, 52(3), 305–316. Huczynski, A., & Buchanan, D. (2001). Organizational Midler, C. (1995). ‘Projectification’ of the firm: The behaviour: An introductory text (4th ed.). Harlow, UK: FT- Renault case. Scandinavian Journal of Management, 11(4), Prentice Hall, Pearson Education Ltd. 363–373. Jessen, S. A. (2002). Business by projects. Miller, R., & Lessard, D. R. (2000). The strategic management Universitetsforlaget AS: Oslo, Norway. of large engineering projects. Boston:Institute of Technology Press. Kharbanda, O., & Pinto, J. (1996). What made Gertie gal- Morris, P. W. G. (1997). The management of projects (2nd lop? Lessons from project failures. New York: Van Nostrand ed.). London: Thomas Telford. Reinhold. Morris, P. W. G. (2002, May). Science, objective knowledge Koskela, L., & Howell, G. (2002, July). The underlying and the theory of project management. Proceedings of the theory of project management is obsolete. Proceedings of PMI Institute of Civil Engineering, 150, 82–90. Research Conference 2002 (pp. 293–301). Newtown Square, Morris P. (2004). Current trends in project and pro- PA: Project Management Institute. gramme management. In Association for Project Management Kreiner, K. (1995). In search of relevance: Project man- Yearbook. High Wycombe, UK: Association for Project agement in drifting environments. Scandinavian Journal of Management (APM).

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 121 Morris, P. W. G., & Hough, G. H. (1987). The anatomy of major Business School, UWE, Bristol, UK. projects. London: Major Projects Association. Thompson, P., & McHugh, D. (2002). Work organisations. Morris, P. W. G., Patel, M. B., & Wearne, S. H. (2000). London: Palgrave. Research into revising the APM project management body of Tjaeder, J., & Thomas, J. (2000, January). On learning and con- knowledge. International Journal of Project Management, 18(3), trol—Competing paradigms or co-existing requirements for managing 155–164. projects in ambiguous situations? Unpublished paper presented to Packendorff, J. (1995). Inquiring into the temporary organi- International Research Network on Organizing by Projects sation: New directions for project management research. (IRNOP) fourth bi-annual conference held in Australia. Scandinavian Journal of Management, 11(4), 319–333. Walta, H. (1995). Dutch project management body of Reed, M. I. (1992). The sociology of organisations: Themes, per- knowledge policy. International Journal of Project Management, spectives, and prospects. Hemel Hempstead, UK: Harvester 13(2), 101–108. Wheatsheaf. Weick, K. (1995). Sensemaking in organisations. Thousand Royal Opera House cancels performance. (1999, November Oaks, CA: Sage. 24). Financial Times, p. 5. Whitehall counts the costs of IT projects. (2000, January 14). Sahlin-Andersson, K., & Söderholm, A. (Eds.). (2002) Beyond Financial Times, p. 13. project management: New perspectives on the temporary-permanent Wideman, R. M. (1995). Criteria for a project management dilemma. Copenhagen: Liber. body of knowledge. International Journal of Project Management, Sauer, C. (1999). Deciding the future for IS failures: Not the 13(2), 71–75. choice you might think. In W. L Currie & B. Galliers (Eds.), Wiig, K. M. (1997). Knowledge management: An introduc- Rethinking management information systems (pp. 279–309). Oxford, tion and perspective. Journal of Knowledge Management, 1(1), 6–14. UK: Oxford University Press. Williams, T. (1995). A classified bibliography of research Silver, C. A. (2000). Where technology and knowledge meet. relating to project risk management. European Journal of Journal of Business Strategy, 21(Nov/Dec.), 28–33. Operational Research, 85(1), 180–188. Söderlund, J. (2004). Building theories of project manage- Williams, T. M. (1999). The need for new paradigm for comment: Past research, questions for the future. International Journal plex projects. International Journal of Project Management, 17(5), of Project Management, 22(3), 183–191. 269–273. Stacey, R. (2001). Complex responsive processes in organisations: Williams, T. (2004, July). Assessing and building on the Learning and knowledge creation. London: Routledge. underlying theory of project management in the light of badly Stallworthy, E. A., & Kharbanda, O. P. (1985). International con- over-run projects. Paper presented at the PMI Research struction and the role of project management. Aldershot, UK: Gower. Conference, London. Standish Group. (1995). Chaos paper. Retrieved March 16th, Winch, G. (1996, September). Thirty years of project man- 2005, from http://www.standishgroup.com/ agement—What have we learned? British Academy of Management Sydow, J., & Staber, U. (2002). The institutional embedded- Conference Proceedings (pp. 8.127–8.145). Birmingham, UK: ness of project networks: The case of content production in Aston Business School. German television. Regional Studies, 36(3), 215–227. Winder, R. (1999, May 16). A grand departure goes uncele- Taggert, W. M., & Silbey, V. (1986). Informational systems: brated. The Independent on Sunday, p. 8. People and computers in organizations. Boston: Allyn & Bacon. Wood, M. (2002). Mind the gap? A processual reconsidera- Thomas, J. (2000). Making sense of project management. In tion of organisational knowledge. Organization, 9(1), 151-171. R. A. Lundin & F. Hartman (Eds.), Projects as business constituents Young, T. (1999). How to be a better project manager. London: and guiding motives. Boston: Kluwer Academic Press. Kogan Page. Thomas, J. (2003, April). Problematizing project management. Young, T. (2003). The handbook of project management – A prac- Paper presented to Making Projects Critical Workshop, Bristol tical guide of effective policies and procedures. London: Kogan Page.

SVETLANA CICMIL, BSc (Civ. Eng), MBA, PhD, is the head of research in operations and information management and director of MBA Programmes at Bristol Business School, University of the West of England, U.K. She has an extensive international teaching, consultancy and industrial experience in the field of project management. Svetlana’s specific research interests lie in the areas of complexity theory and sociology of project based work, project management education, and the development of project management skills, competencies, and knowledge. She is also interested in the global aspects of management education, and the international transfer of management knowledge. Svetlana is actively involved in Project Management Research Networks, regularly speaks at conferences and has published widely. Recent articles written by Svetlana have appeared in International Journal of Project Management, Building Research and Information, and International Journal of Diversity in Organisations, Communities, and Nations. In collaboration with Dr. Hodgson, Svetlana published an influential book Making Projects Critical (Palgrave), drawing on a series of distinctive workshops with the same name with contributions from a range of international researchers.

DAMIAN HODGSON BA (HONS), MA, PhD is senior lecturer in organisational analysis at Manchester Business School, University of Manchester. His key research theme at present is the phenomenon of projects in work organisations, encompassing an analysis of the emergence of project-based organisations, the widespread adoption of project working across a range of industries, and ongoing efforts to institutionalise project management as a professional discipline. In particular, he is keen to promote a critical analysis of project management knowledge, in terms of its epistemology, ontology and related core assumptions. Damian is also co-investigator on the EPSRC-funded network entitled “Rethinking Project Management” (www.rethinkingpm.org.uk). He has presented work at a variety of international conferences including the Academy of Management, the Critical Management Studies Conference, the British Academy of Management, and EGOS, and has published in a range of journals including the Journal of Management Studies, Organization, Management Decision and Gender, Work and Organisation.

122 AUGUST 2006 PROJECT MANAGEMENT JOURNAL Notes for Authors

SCOPE SHORT ITEMS applications and/or Rich Text Format (RTF). It is Project Management Journal is the professional Short items do not need rigorous academic essential that the name and version of the word journal of the Project Management Institute. The scrutiny and are not refereed. Upon receipt, how- processing program and format of the text files are mission of the Journal is to advance the state of ever, these items become the copyright property clearly indicated (example: Word for Windows the art of the knowledge of project and program of PMI. 2000 doc). The electronic version should only be management. The Journal presents useful infor- Opinion presents thoughtful discussion of sent with the final accepted version of the paper to mation on both theory and practice in the field project management issues. the Editor. NOTE: The hard copy and electronic of project management. Authors are encouraged Correspondence pertains to the project and files must match exactly. to submit the following types of original manu- program management profession, including ref- Upon acceptance of the manuscript for scripts: descriptions of innovative practices; sum- erences to literature, practice, and scholarship as publishing, authors will also be asked to provide maries of research results; reviews of current well as discussion and replies related to articles illustrations placed or embedded within their literature; surveys of current practices; critical published in the Journal. chosen word processing program. If this isn’t analyses of concepts, theories, or practices; devel- Book Reviews express opinions about books possible, please submit illustrations in their opments of concepts, theories, or practices; related to the project management profession, or native programs. Be sure to include a hard copy analyses of failure. Manuscript length should not about general management or technical books as well. PMI now recreates all illustrations, fig- exceed 12,000 words. The selection of manu- that cover topics of particular value to the project ures and tables electronically for publication. By scripts for publication is based on the extent to manager. doing so, the publication is ensured a consistent which they advance the knowledge and under- Calendar of Events offers notices of forth- look thoughout. Although this makes electronic standing of project management. PMI neither coming meetings and calls for papers. versions of illustrations less important, a hard approves nor disapproves of any data, claims, copy becomes crucial for re-creation purposes. opinions, or conclusions presented. SUBMISSIONS Contact PMI’s Publishing Department for fur- All manuscripts must be submitted electronical- ther details. MANUSCRIPT REVIEW ly either by e-mail to [email protected] Project Management Journal uses a double-blind or on CD and sent to: Project Management STYLE OF TEXT review process. The first review of every manu- Journal, Attn: Natasha Pollard, 4 Campus Blvd., You should write in clear and concise English. script is performed by two anonymous referees Newtown Square, PA 19073 USA. If you submit Spelling should follow Webster’s New World (usually members of the Editorial Review your manuscript on CD, please include a print- Dictionary. Authors whose native language is not Board). The manuscript is then either accepted, out of the manuscript, including all tables and English are assured that in-house editorial atten- rejected, or returned to the author for revision figures, on 8-1/2 x 11 inch paper, double spaced tion to their manuscript will improve clarity and (with reviewer comments furnished to the throughout, and printed on one side only. acceptability to readers. For questions regarding author). Revised manuscripts are sent to the Manuscripts should include the following in the style and format of text, refer to the Publication Editor, who makes a final disposition in con- order listed: Manual of the American Psychological Association, sultation with the Publisher. The Journal strives A title page that includes the title of the Fifth Edition. to respond to all authors within three months manuscript and each author’s name, affiliation, of the date the manuscript is received at the mailing address, and phone, fax, and e-mail REFERENCES PMI Publishing Department. Accepted manu- address. Correspondence will be directed only to For questions regarding reference format, refer scripts are subject to editorial changes. The the first author listed. to the Publication Manual of the American author is solely responsible for all statements An abstract of 100 words or less that out- Psychological Association, Fifth Edition. made in the manuscript, including editorial lines the purpose, scope and conclusions of the References used in the text should be identified changes. manuscript, and selected keywords. by author name and publication date in paren- Text (use headings and no more than two theses, e.g., (Cleland & King, 1983), and listed ORIGINAL PUBLICATION levels of subheadings). To permit objective alphabetically at the end of the manuscript. It is the policy of PMI to be the sole, original pub- reviews by two referees, the abstract and first Page numbers should be cited for all quota- lisher of manuscripts. Manuscripts that have been page of the text should not reveal the authors tions. Follow the format example shown below: submitted simultaneously to other magazines or and/or affiliations, but only the manuscript title. Baker, Bud. (1993). The project journals will be rejected outright and will not be References. manager and the media: Some les- reconsidered. Republication of a manuscript, Illustrations and Tables. These should be titled, sons from the stealth bomber pro- possibly revised, that has been disseminated via numbered (in arabic numerals and captions), and gram. Project Management Journal, 24 conference proceedings or newsletter is permitted each on a separate sheet, and the preferred location (3), 11–14. if the Editor judges there are significant benefits indicated within the body of the text. Cleland, David I., & King, to be gained from publication. Biographical details of each author. William R. (1983). Systems analysis Upon manuscript acceptance, authors must also and project management. New York: COPYRIGHT provide a black-and-white passport-style photo- McGraw-Hill. Upon acceptance of a manuscript, authors will graph and a signed copyright agreement. Hartley, John R. (1992). be asked to transfer copyright of the article to the Concurrent engineering. Cambridge, publisher. This transfer will ensure the widest COMPUTER-GENERATED MA: Productivity Press. possible dissemination of information. This TEXT AND ILLUSTRATIONS Please ensure that references are complete, that transfer of copyright enables PMI to protect the Authors are requested to submit the final text they include, where relevant, author’s name, copyrighted material for the authors, but does and illustrations via e-mail or on CD. As with the article or book title, volume and issue number, not relinquish the author’s proprietary rights. requirements for manuscript submission, the publisher, date and page reference. The copyright transfer gives PMI the exclusive main text, list of references, table and illustration The use of page footnotes should be kept rights to republish or reprint the manuscript in captions, and author biographies should be to a minimum. Footnotes should be numbered any form or medium as well as to grant or refuse stored in separate text files with clearly identifi- consecutively and listed at the end of the text permission to third parties to republish all or able file names. Keep the layout of the text as as endnotes. parts of the manuscript. simple as possible and save text in its original

AUGUST 2006 PROJECT MANAGEMENT JOURNAL 123 KEYWORDS Keywords categorize your manuscript. They cover project management methodologies and processes, tools and techniques, PMBOK® Guide knowledge areas, industries, types of projects, geography. Please list three or four keywords that best categorize your manuscript. Choose from the following list of suggested keywords (this is not a comprehensive list) or you may use your own.

Accounting Multiproject Planning CHECKLIST Activity Duration Estimating Negotiating Manuscript via e-mail or on CD Agriculture Networking 100-word abstract Arrow Diagramming Method New Product Development Illustrations Baselines Organizational Planning Author biographies Benchmarking Organizational Structure Black and white author photographs Benefit/Cost Analysis Parametric Modeling (upon acceptance) Budgeting Performance Reporting Signed copyright agreement Change Control Pharmaceuticals (upon acceptance) Communications Management Procurement Management REVISIONS Concurrent Engineering Productivity Correspondence and files for revision will Configuration Management Project Life Cycle be sent to the first-named author unless Conflict Resolution Project Management Software otherwise indicated. Copyediting of Constraints Project Plan Development manuscripts is performed by PMI staff. Construction Quality Assurance The authors are asked to check edited files for Contingency Planning Quality Management typographical errors and to answer queries Contract Closeout Reengineering from editors. To improve publication times it Cost Estimating Resource Planning is important that revised files be returned Cost Management Responsibility within three days. Excessive turnaround time Critical Path Risk Management may jeopardize publication of papers. Delegation Risk Response Development Deliverables Schedule Development COPIES AND REPRINTS Design Schedule Control Authors will receive 10 copies of the Journal Documentation Scope Management free of charge. Additional copies of the Journal Earned Value Scope Definition and/or article reprints can be ordered at any Engineering Scope Change Control time from PMI. Environment Simulation Project Management Institute Estimating Staff Acquisition Publishing Department Fast-Tracking Stakeholders 4 Campus Blvd. Feedback Standards Newtown Square, PA 19073 USA Finance Statistical Sampling Tel: 610/356-4600 ext. 5016 Float Team Development Fax: 610/355-1633 Funding Time Management E-mail: [email protected] Human Resource Management Tools Information Systems Training Integration Management Transportation Variance Large Project Utilities Leadership Virtual Organization Life-cycle Costing Work Breakdown Structure Manufacturing Work Packages Management Skills Matrix Organization Milestones Mitigation Monte Carlo Analysis

Index of Advertisers

Soundview Executive Book Summaries ...... C2

The Standards for Program and Portfolio Management ...... C3

PMI Global Congress Latin America ...... C4

124 AUGUST 2006 PROJECT MANAGEMENT JOURNAL PROJECT

PROGRAM MANAGE SUCCESS.

Organizations invest in their future by initiating projects, programs and portfolios as a means of achieving organizational success. The utilization of program and portfolio management processes and skilled program and PORTFOLIO portfolio managers continues to rise. Recognizing this development, PMI introduces The Standard for Program Management and The Standard for Portfolio Management.

Learn how to optimize program Portfolio management is gaining components and organizational favor as a means of helping schedules and resources organizations focus not only for success. The Standard on “doing work right” but also on for Program Management “doing the right work.” The Standard describes the processes for Portfolio Management provides and provides guidance for methods to take a comprehensive managing multiple projects and view of portfolios, group them for non-project activities within a the most effective management program environment. and ensure all components are aligned strategically. PMI Member Price : $38.95 (US) Price : $46.95 (US) PMI Member Price : $38.95 (US) Price : $46.95 (US)

Take the next step toward managing the success of your organization. Visit the PMI Bookstore (www.pmibookstore.org) for more information and to order your copy today.

Global Operations Center Regional Service Centre Regional Service Centre Four Campus Boulevard Europe-Middle East-Africa (EMEA) Asia Paciﬁc Newtown Square, PA 19073-3299 USA Avenue de Tervueren 300 73 Bukit Timah Road Tel: +1-610-356-4600 B-1150 Brussels, Belgium #03-01 Rex House Making project management indispensable for business results.® Fax: +1-610-356-4647 Tel: +32-2-743 15 73 Singapore, 229832 E-mail: [email protected] Fax: +32-2-743 15 50 Tel: +65 6330 6733 Internet: www.pmi.org E-mail: [email protected] Fax: +65 6336 2263 E-mail: asiapaciﬁ[email protected] ©2006 Project Management Institute, Inc. All rights reserved. “PMI”, the PMI logo and “Making project management indispensable for business results” are marks of Project Management Institute, Inc. 045-009-2006 (03-06)