Corporate Brand Impact on Sales / Revenue Per Share

University of South Florida Scholar Commons

Graduate Theses and Dissertations Graduate School

November 2020

Brad A. Puckey University of South Florida

Follow this and additional works at: https://scholarcommons.usf.edu/etd

Part of the Business Administration, Management, and Operations Commons, and the Marketing Commons

Scholar Commons Citation Puckey, Brad A., "Corporate Brand Impact on Sales / Revenue Per Share" (2020). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/8582

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected].

Corporate Brand Impact on Sales / Revenue Per Share

Brad A. Puckey

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Business Administration Muma College of Business University of South Florida

Co-Major Professor: Loran Jarrett, D.B.A. Co Major Professor: Paul Solomon, Ph.D. Jung Chul Park, Ph.D. Joann Quinn, Ph.D.

Date of Approval: October 30, 2020

Keywords: Branding, Quantitative, CHAID, Branding Value

DEDICATION

To John B. “Jack” Frey; he was a brilliant colleague, a wonderful mentor, and an even better friend. You taught me how to put education to practice with knowledge and wisdom as well as how to turn numbers into intelligence. You are dearly missed.

ACKNOWLEDGMENTS

The completion of this dissertation would not have been possible without the contribution of many wonderful people. I’d like to start by thanking my committee members: Defense Chair,

Dr. Jung Chul Park, co-chairs Dr. Joran Jarrett and Dr. Paul Soloman for all their wisdom and advice throughout the process and committee member Dr. Joann Quinn for her wisdom and guidance.

I’d also like to acknowledge and thank the entire 2020 cohort of fellow doctoral students for sharing their experiences, knowledge, and general good wit during the entire course of instruction. I’d especially like to thank the cohort members on my committee, Mark Mattia,

Michael Summers, and Toufic “Tom” Chebib. Your constant encouragement and assistance helped put this work over the line.

Thank you to Hampton Bridwell at Tenet Partners for sharing the CoreBrand Index® database used in this research. A sincere thank you to Kellan Williams, my analytical partner in this analysis; his skill and wisdom helped turn our pile of data into a story.

A special thank you to my Uncle and Aunt, Dr. James and Evelyn Gregory. Your help in every aspect of this process made it possible. I simply could not have done this without your support. Your encouragement, advice and counsel, friendly ear when I was about to come off the rails, and material support made this happen. Thank you!

Countless others deserve my thanks, but I have run out of room. I am truly blessed and say thanks to all of you!

TABLE OF CONTENTS

List of Tables ...... iii

List of Figures ...... iv

Abstract ...... v

Chapter One: Introduction and Background ...... 1 Introduction ...... 1 Stating the Problem ...... 2 Research Question and Units of Analysis ...... 3 Substantive Focus ...... 3 Specific Theoretical Discussions Addressed ...... 5 Concepts and Definitions ...... 7 Summary and Organization ...... 8

Chapter Two: Literature Review ...... 9 The Corporate Brand Impact ...... 9 Literature Search Strategy...... 10 Analysis Process ...... 10 Types of Brand Measurement Methodologies ...... 11 Theory of Intangible Capital ...... 13 Corporate Brand Construct ...... 14 Marketing Theory ...... 15 Brand Theory ...... 16 Product Market Model ...... 16 Gaps in the Literature ...... 17 Hypotheses ...... 17

Chapter Three: Methodology ...... 19 Purpose...... 19 Sales/Revenue-per-Share Explained...... 19 Practical Business Application of the Model ...... 20 Research Design and Data Collection ...... 20 Key Descriptive Statistics ...... 25 Key ...... 26 CHAID Model Overview ...... 27

Chapter Four: Results of the Study ...... 29 Analysis ...... 29

Data Preparation ...... 29 Initial Data Exploration – Univariate Test ...... 30 Correlation Analysis ...... 32 Hypothesis Testing ...... 34 CHAID Analysis – Limited Micro Model ...... 35 CHAID Analysis – Full Macro Model ...... 39 Marco Model Results ...... 42 Additional Models ...... 43 Consumer Staples Sector ...... 43 Industrials Sector ...... 44 Technology Sector ...... 46 Comparing the Sector Models ...... 48 Industry Analysis ...... 49

Chapter Five: Summary and Discussion ...... 50 Summary of the Problem ...... 51 Research Question ...... 51 Hypotheses ...... 51 Summary of the Methodology ...... 52 Summary of the Results ...... 52 Contributions to Knowledge ...... 53 Practitioner Contribution ...... 54 Study Limitations and Future Research ...... 56 Conclusions ...... 57

References ...... 60

Appendix A: Descriptive Statistics ...... 64

Appendix B: Correlation Analysis ...... 70

Appendix C: Sector CHAID Decision Trees ...... 75

Appendix D: Permission Letter ...... 77

About the Author ...... End Page

LIST OF TABLES

Table 1: Quintile Analysis of Key Data for Companies in the Database ...... 31

Table A1: List of Variables and Descriptive Statistics ...... 64

Table A2: 1-Year Sales / Revenue Per Share, % Growth Correlations ...... 70

Table A3: 3-Year Sales / Revenue Per Share, % Growth Correlations ...... 71

Table A4: 5-Year Sales / Revenue Per Share, % Growth Correlations ...... 73

iii

LIST OF FIGURES

Figure 1: Impact of Industry and Corporate Brand Factors on 5-yr Sales/ Revenue Per Share % Growth Rate ...... 35

Figure 2: CHAID Decision Tree Macro Analysis, Industry and Corporate Brand Model ...... 37

Figure 3: Industries Included in Nodes 1-4 of the CHAID Analysis ...... 38

Figure 4: Predictor Importance in Full Macro Model ...... 40

Figure 5: CHAID Full Macro Model Decision Tree ...... 40

Figure 6: Industries Included in Nodes 6, 7, and 8 ...... 41

Figure 7: Industries Included in Nodes 9, 10, 11, 12, and 13 ...... 41

Figure 8: Industries Included in Nodes 14, 15, and 16 ...... 42

Figure 9: Results of the Macro Model ...... 42

Figure 10: Predictor Importance Consumer Staples Sector ...... 44

Figure 11: Results of Consumer Staples Sector Model ...... 44

Figure 12: Predictor Importance Industrials Sector ...... 45

Figure 13: Results of Industrial Sector Model ...... 46

Figure 14: Predictor Importance Technology Sector ...... 47

Figure 15: Results of Technology Sector Model...... 48

Figure A1: CHAID Decision Tree: Consumer Staples Sector ...... 75

Figure A2: CHAID Decision Tree: Industrials Sector ...... 75

Figure A3: CHAID Decision Tree: Technology Sector ...... 76

ABSTRACT

The purpose of this dissertation is to identify the linkage between the corporate brand and sales / revenue per share growth. Utilizing a unique database, the CoreBrand® Index, this study is a quantitative analysis of the interaction between the corporate brand and financial data. A relationship between corporate brand and financial data has been suspected by practitioners; however, it has never been proven. For the first time, quantitative analysis is conducted in this dissertation using this unique dataset.

This study is intended to help those responsible for managing corporate brands to better understand their impacts and benefits. The more that value creation can be understood and explained, the more likely that corporate leadership will view expenditures as investments rather than simply expenses. This study is intended to help demonstrate how corporate brands can be used to create value.

A chi-squared automatic interaction detector (CHAID) analysis was performed to predict whether a company would have high or low sales / revenue per share growth. This analysis was done on the macro and sector level with results exceeding 80% accuracy. This analysis can be used to help practitioners understand and predict the corporate brand’s impact on business results and help them justify and allocate resources to manage this asset.

This work can help corporations to justify and manage their corporate brand budgets and provide a basis for accountability. The corporate brand’s impact on sales / revenue per share is but one small piece of intangible capital that can be understood and managed. The more elements

v that can be uncovered and managed, the better companies can predict and plan for their financial growth and reduce the uncertainty surrounding their businesses.

CHAPTER ONE:

INTRODUCTION AND BACKGROUND

Introduction

Branding has been a topic of study by marketers since the late 19th century. The branding concept is that by enriching the awareness, reputation, and perception of a product, higher customer loyalty and increased sales can be obtained (Peng & Li, 2019).

The corporate brand, while having been studied for decades, is a newer concept. Around the 1970s, advertisers began creating a personality for their companies to create traction for their product brands and to make product launches more successful (Holland, 2017). Traditional beliefs are that product branding can increase revenue and corporate brand can increase shareholder value and goodwill (Gregory & Wiechmann, 1991; Puckey, 2012). There has always been a general belief that corporate branding can improve sales, but no macro-level empirical proof has been obtained. This dissertation uses existing corporate brand data provided by Tenet

Partners, in conjunction with financial statistics and company-paid media investment data, to seek empirical evidence of corporate brand impact on sales revenue growth. Tenet Partners is a leading corporate brand consulting agency headquartered in New York, NY, formed by the merger of Brandlogic and CoreBrand in 2014.

Gregory’s 2018 paper, “Intangible Capital: Culture of Innovation and its Impact on the

Cash Flow Multiple,” examined the corporate brand and its impact on the cash flow multiple as a component of intangible capital. Further, his work asserted that intangible capital is composed of

1 many layers, and the more fully we can understand those layers, the more complete our understanding of intangible capital would be (Gregory, 2018). The present study extends

Gregory’s work by attempting to understand another component of the corporate brand’s contribution to intangible capital with the goal of expanding our understanding of it.

Given the longitudinal nature of the data available, this dissertation examined how the corporate brand, in its totality and components, changes over time and impacts changes in a company’s revenue per share. Rather than serving as a precise predictive model, the underlying objective was to understand the impact and what magnitude it achieves. Additionally, this research evaluated how the different components of corporate brand have varying impact on sales/revenue per share. Future research on product branding can use this research to determine how corporate and product brands may work together to create value and to predict and model outcomes.

The outcome of this dissertation will help corporate communications executives better understand how the corporate brand contributes to value creation. This increased understanding can offer additional evidence to prove the value to senior company leaders and to make better decisions when employing the corporate brand for use. Better decisions will help businesses secure resources and make strategic and tactical communications decisions.

Stating the Problem

Intuitively, marketers and senior executives believe that the corporate brand contributes to a company’s business performance. However, the questions raised since the 1970s are: How much? How long does it take? How much does it cost? Answers to these questions have been elusive because there has been no quantitative tangible measure of corporate brand performance on a large enough scale. Without answers to these questions, senior executives can be reluctant

2 to adequately fund an effort that they have difficulty measuring its performance and tracking accountability. The goal of this research is to answer those questions in an effort to make the corporate brand more tangible to those outside of corporate marketing and help senior executives better understand the value of investing in the corporate brand.

Research Question and Units of Analysis

The research question was: How does the performance of the corporate brand impact sales / revenue per share?

This question is raised and measured at the corporate level. By measuring the strength and growth in the corporate brand, in conjunction with strength and growth in corporate sales/revenue per share, quantitative analysis of the impact can be performed. This study examines archival data collected on the corporate brand of 621 corporate brands from 2011 to

2016. Also, it includes total paid media spending annually for each of those companies as well as corporate level financial data, using sales revenue per share as the dependent variable.

This dissertation focuses on the corporate brand as a component of intangible capital.

Intangible capital is composed of multiple layers of value; this research attempts to identify one of those layers. It looks for how the corporate brand, in conjunction with other factors, can impact a company’s sales / revenue per share. Differences in industry and sectors as well as brand architecture are examined to help identify how the impact of corporate brand varies in different conditions. The learning generated is intended to help corporate brand managers make better informed decisions and manage expectations for their efforts.

Substantive Focus

The data used is archival research collected by Tenet Partners for clients’ use. The companies studied are among the largest global companies in the world, including Apple,

Microsoft, GE, and many others. The data is collected in the United States and based on the company’s corporate brand, which is measured by evaluating respondents’ familiarity with a brand and, if familiar, evaluating their perceptions across several key favorability attributes. As the centerpiece of the research, this data is used as a quantitative representation of the corporate brand to be used in conjunction with fundamental financial variables and paid media spending.

This quantitative research study is designed to isolate and identify the impact of a company’s corporate brand on sales/revenue per share growth.

The corporate brand data consists of measures of familiarity and three favorability attributes gathered in Tenet Partners’ CoreBrand® Index (CBI). Tenet Partners is a corporate brand consulting firm based in New York, NY. The favorability attributes consist of overall reputation, perception of management, and investment potential. These attributes are averaged to create a Favorability score. All corporate brand measures are expressed as scores from 0–100. To create a single measure of the corporate brand (known as BrandPower), Familiarity and

Favorability are multiplied together and then multiplied by 0.01 to reduce the result to a 100- point scale. This data was continuously collected throughout the year via telephone interviews averaging 10 – 12 minutes each.

Ultimately, sales / revenue per share was used as the dependent variable with the goal of identifying if stronger corporate brands result in higher sales revenue. Because of the time-series nature of this data from all the sources, the research was focused on change over time. This focus allowed the evaluation of growth rates, the impact of the level of brand and its changes, and how things like paid media investment may influence results.

Ultimately, it is hoped that this study can change the way the corporate brand is perceived. The sought-after outcome is for senior corporate leadership to evolve its thinking

4 relative to the value of the corporate brand. Over time and with the right evidence, it is desirable for senior leadership to view the corporate brand as an asset that has value and should be invested in, rather than an expensive nuisance that consumes resources with dubious return.

Specific Theoretical Discussions Addressed

Primary to this research is the theory of intangible capital, as discussed by Gregory

(2018). In that work, the author theorizes that intangible capital is comprised of many components, including product brand, corporate brand, customer loyalty, and other factors. He also shows that, while it is not Generally Accepted Accounting Principles (GAAP) compliant nor on the balance sheet, the corporate brand contributes value to an organization, i.e., value that can be measured and managed for company growth. Gregory’s work concentrates on the corporate brand’s contribution to market cap as measured through the cash flow multiple (stock price/cash flow per share). As an extension of his analysis, the present study should be viewed as complementary to Gregory’s work. By understanding how the corporate brand impacts sales/revenue per share as well as cash flow multiple--as demonstrated in Gregory’s work--a more complete picture emerges of how the corporate brand contributes to total value creation.

The present study reveals the sales/ revenue component of intangible capital.

In the marketing context, institutional theory is another useful theoretical base for the present study. Marketing helps communications executives direct their messages towards markets and consumers, rather than satisfy institutional pressures and inter-organizational relations (Slimane et al., 2019). This clarity allows marketing executives to convey their messages to clients to sell product, rather than simply please internal audiences but forfeit any inroads on sales. The present study extends this theory by emphasizing that the corporate brand

5 can be communicated much like a product brand to increase its value (Gregory & Weichmann,

1991).

One issue of concern is that a specific theory for the construct of a brand name element is nonexistent (Round & Roper, 2012). Round and Roper identified that when brand equity is measured in the literature, it typically focuses on measures of brand awareness and brand association. These measures assess the respondent’s knowledge of a brand and his/her perceptions of it on various traits of that brand. This approach is similar to the measures used in the present study to gage equity in the corporate brand. These measures are Tenet Partners’ data on corporate brand Familiarity and Favorability. By comparison, both constructs are measures of the size of the brand, brand awareness and Familiarity, and brand perception, brand association and Favorability. In sum, I examined the brand’s mass and quality because these measures offer a quantitative representation of the brand, which is an emotional concept that can be difficult to infer.

Traditional approaches to corporate brand assert that the corporate brand is an entity that is created and carefully managed by corporate leadership (Balmer, 2001). Meanwhile, Social constructionists’ approaches assert that the corporate brand is a function of the social interaction between the organization and the environment in which it operates (Leitch & Richardson, 2003).

Often, these two concepts are presented as differing points-of-view regarding corporate branding

(Melewar et al., 2012). It is likely that both approaches are concepts that are correct and the weight of influence of each concept on a corporate brand would depend on the brand in question and the way in which the company operates.

Concepts and Definitions

The construct used to measure the strength of the corporate brand in this research is

BrandPower, which is the result of survey data collected by Tenet Partners. Quantitative measures of a corporate brand’s Familiarity and Favorability (as measured by overall reputation, perception of management, and investment potential) were collected to be used as input variables in any empirical research where the corporate brand could be considered an independent or dependent variable. BrandPower represents the interaction between Familiarity and Favorability as represented by a single number for each company at a point in time. This data was collected continuously each year via telephone interviews.

The data is intended to be a quantitative representation of the corporate brand that can interact with other quantitative performance measures of company performance, such as stock price, sales, product volume, and other measures. While BrandPower is not actually the corporate brand, which is an emotional concept defined by the individual, it is a viable quantitative representation of the corporate brand that can be used to measure what impacts corporate brand performance and how it can impact business results.

Conducting the BrandPower research is critical to building accountability for corporate brand performance within organizations. By creating a quantitative measure of the brand, researchers can evaluate the corporate brand’s performance and impact among other forms of corporate data, such as financial data and marketing spend data. Most managers intuitively understand the impact that corporate brand can have and the results that can be produced

(Puckey, 2012). However, the magnitude of the impact needs to be defined and measured before true accountability can be achieved.

The concept of intangible capital also is key to this research. There are many components of intangible capital; these components need to be measured and valued to understand their contribution. These measurements and valuations are not GAAP compliant and do not end up on the balance sheet; however, this does not diminish their importance (Gregory, 2018). They are operational factors that reflect the health of the business and help executives diagnose procedures that may be necessary if the company is functioning poorly.

Summary and Organization

This introduction provides a high-level overview of the problem that faces communications executives and the difficulty they have in demonstrating the value of their work.

That difficulty can impede their ability to secure the resources needed to adequately perform their job. This dissertation introduces a way to measure the corporate brand, identify its impact, and provide a conduit through which marketing executives can be recognized for contributing to their companies’ success. This research is not an attempt to change GAAP principles, but rather a practical way to measure the corporate brand’s contribution to a company’s performance.

Specifically, this dissertation utilizes the CoreBrand Index historical data to identify the corporate brand’s contribution to a company’s sales/revenue per share growth. The measures of familiarity, overall reputation, perception of management, investment potential, and culture of innovation combine to give a measure of the corporate brand known as “BrandPower.” These measures provide a quantitative measure of the corporate brand that can be evaluated in the context of quantitative financial measures and communications efforts.

CHAPTER TWO:

LITERATURE REVIEW

The Corporate Brand Impact

According to Clancy and Krieg’s (2007) book, Your Gut Is Still Not Smarter Than Your

Head: How Disciplined, Fact-based Marketing Can Drive Extraordinary Growth and Profits, the marketing industry has been driven by gut instinct for decades. Nobody understood what worked or why, and executives rarely had to defend their budgets. Things have changed, and marketers are now accountable to the bottom line like everyone else. Research and scientific modeling are now required to justify expenditures and identify expected returns. This approach allows marketers to change from thinking with their gut to using state-of-the-art analytics to help them think with their heads (Clancy & Krieg, 2007).

The corporate brand is a unique asset for each company in operation. It is different than the product brand in that it is the brand of the company, not the brand of an individual product or line of products. The question posed in this research is: “Does the corporate brand contribute to sales per share growth, and if so, how does it contribute?”. To better understand this relationship,

Tenet Partners CoreBrand Index® was utilized. The primary measure of the corporate brand is

BrandPower, which is a measure of Familiarity and Favorability. Familiarity relates to public recognition of the brand. Favorability can be further broken into Overall Reputation, Perception of Management, and Investment Potential. The goal of this research was to understand how the corporate brand, represented by BrandPower, impacts sales- per-share growth and what role is

9 played by these components of BrandPower. This analysis used time-series data across hundreds of companies, representing 10 sectors and 52 industries to explore this relationship.

Literature Search Strategy

For the literature review, I utilized the University of South Florida (USF) library and my literature collection on the topic of branding. I examined the topic of branding from a consumer- and corporate perspective. Theory of brand and brand architecture were also examined. Also,

Google Scholar was utilized; search terms included brand architecture, brand theory, consumer branding, corporate brands, corporate vs. product brands, marketing theory, and brand familiarity and favorability.

Analysis Process

Before discussion of the corporate brand and its role, the analytical framework should be briefly discussed. This study used a data driven quantitative analysis. These types of analyses must be methodical in their process. The researcher cannot “make it up as s/he goes.” It requires forethought and process application. In his textbook, Basic Econometrics, Gujarati (1995) outlines the Methodology of Econometrics, which follows the steps listed below for conducting this type of quantitative research:

1. Statement of theory or hypothesis

2. Specification of the mathematical model of the theory

3. Specification of the econometric model of the theory

4. Obtaining the data

5. Estimation of the parameters of the economic model

6. Hypothesis testing

7. Forecasting or prediction

8. Using the model for control or policy purposes.”

(Gujarati, 1995, p. 3).

While the modeling in this dissertation is a statistical model rather than an econometric model, these steps are a good basic procedural overview. The research in this dissertation did not follow this process to the letter, but the process was loosely adhered to for this study.

Types of Brand Measurement Methodologies

Keller’s (2001) model for building a strong brand includes four steps:

• Establishing the proper brand identity

• Creating appropriate brand-meaning through strong and positive associations

• Eliciting positive responses

• Establishing loyal customer relationships

When a marketer can do these steps, positive benefits are gained. As a result of accomplishing the above four steps, six brand-building blocks are achieved:

• Brand salience

• Brand Performance

• Brand imagery

• Brand judgments

• Brand feelings

• Brand resonance

Establishing these principles helps a marketer establish market-research priorities, set strategic direction, and inform brand-related decisions.

Fischer (2007) asserts that measures of financial brand valuation must adhere to modern accounting standards. As such, they should be:

• Future oriented

• Objective

• Complete

• Comparable

• Simple

• Cost effective

Additionally, they should be adaptable to specific conditions of the organization, such as reporting and planning.

Mintz and Currim (2003) acknowledge that marketers and senior management do not speak the same language. In an effort to increase accountability, the authors mention that the

Journal of Marketing, the Marketing Science Institute (MSI), and the Institute for the Study of

Business Markets (ISBM) recommend the use of marketing metrics that link marketing and marketing-mix inputs with financial business outputs to define success. This approach was influential for the research in this dissertation as it seeks to measure how the corporate brand can impact sales/revenue-per-share growth. The corporate brand is not expected to be the primary driver of sales/revenue growth; however, it must be tied to the success metrics used by senior management to have value to them.

A small sample of some types of measurement activities and outcomes mentioned by

Mintz and Currim (2003) is outlined below:

• Market share (units or dollars) leading to profit margin

• Awareness (product or brand) leading to return-on-investment

• Satisfaction (product or brand) leading to return-on-sales

• Likeability (product or brand) leading to return-on-marketing-investment

• Preference (product or brand) leading to net present value

• Willingness to recommend (product or brand) leading to economic value added.

The wide array of measurement and outcome methodologies highlights the importance of understanding the brand being measured and the outcomes that are important to senior management. It also highlights key components to measurement that must be understood. There are two key reasons to do brand measurement: 1) to influence marketing activities to optimize outcomes and 2) to convince leadership of the need to support communications efforts. The measurement activities in this research study were intended to satisfy the latter. For that reason, the measurement must be relevant and matter to senior management.

Theory of Intangible Capital

The Theory of Intangible Capital was advanced by Gregory in his 2018 dissertation that measured the impact of the Culture of Innovation attribute on the cash-flow multiple. Gregory identifies the Culture of Innovation as an added contributor to the valuation of a company. The key idea in the Theory of Intangible Capital is that a company creates value in many forms; the more value we can quantitatively identify, the better we can measure and manage value creation.

This dissertation identifies yet another layer of value that has not yet been revealed. The understanding is this additional layer will never be a GAAP-recognized value measure that is on the balance sheet, but it is a measurement system that can be used to help communications executives better identify value-creation to help better manage their assets and do their jobs. This dissertation is intended to continue the understanding of the value of the corporate brand. This study does not identify all of the value of corporate brand; rather, it highlights one additional component of value. Future research will identify other elements of value and possibly combine them to create a more complete picture.

A further reason to better understand intangible capital was highlighted by Haigh (2017) in his response to a critique by MARKABLES. He states that, in the United States and American companies, a tendency exists to understate the value of identifiable, intangible assets. Further, there is a trend towards over-reliance on goodwill, leading to an overvaluation of the brand during acquisitions, something that may not be recognizable in the immediate future. This overvaluation provides further evidence for the need to identify and understand intangible capital. In sum, there is reason to recognize brand value, not only for the optimization of marketing activities, but also for recognizing brand value during acquisitions.

Corporate Brand Construct

A specific theory for the construct of a brand-name element has not existed (Round &

Roper, 2012). According to Round and Roper, when brand equity is measured, it typically focuses on brand awareness and brand association. These measures assess the respondent’s knowledge of a brand and his or her perceptions of it on various traits of that brand. This assessment is similar to the measures that I used to measure equity in the corporate brand. These measures are from Tenet Partners’ data on corporate-brand Familiarity and Favorability. By comparison, both constructs are measures of the size of the brand, brand awareness and

Familiarity, and brand perception, brand association and Favorability. In other words, they are the brand’s mass and quality. These measures offer a quantitative construct of the brand (i.e., an emotional concept that can be difficult to articulate).

Aaker (2004) identifies the corporate brand as the brand that defines the organization, stands behind the offering, and is defined by organization associations. The corporate brand may or may not be relevant to product brands. The corporate brand defines the personality and characteristics of the company as it relates to other organizations. This definition is true of the

BrandPower data studied in my research. This data identifies the strength of the organizational brand and measures it consistently across companies, industries, and sectors, which is why

BrandPower and its components are an appropriate, discreet, and quantitative representation of the corporate brand to use in statistical modeling.

Marketing Theory

Before an understanding of the corporate brand’s impact on sales-per-share can be developed, key building blocks must be understood. A logical starting point is marketing theory.

In his work at University St. Gallen, Ludicke (2006) identified marketing as a social system, with brand being the main pressure point. The social system is based on communications systems that have brand as the focus of those communications.

Marketing helps communications executives direct their communications towards markets and consumers, rather than satisfy institutional pressures and inter-organizational relations (Slimane et al., 2019). This clarity allows marketing executives to convey their messages to clients to sell product, rather than simply please internal audiences but not make any inroads on sales.

It is important for marketers to abide by legal and ethical considerations as they pursue their practice. Regulatory bodies, such as the Federal Communications Commission (FCC), provide legal oversite to communications activities (Gazley et al., 2016). Many recent scandals from Enron to B.P.’s Deepwater Horizon have eroded the trust between corporate leadership and the consuming public. This harm extends beyond trust and bleeds into significant financial consequences for the stakeholders.

Brand Theory

Chandler and Owen (2002) indicate that brands affect customer behaviors by transmitting signals that provide symbolic meanings that appeal to customers’ wants and needs. Brands like iPhone, Google, and Uber are a few such brands. When they are mentioned, no one needs to tell you what services are provided; you simply know.

Product Market Model

According to Ailawadi et al., (2003), brand equity measures exist in three distinct markets and consist of various outcomes. These are the consumer mind-set, the product market outcome, and the financial market outcome. The consumer mind-set is where the customer is the source of brand equity. It is based on measures of awareness, attitudes, associations, attachments, and loyalties that customers associate with brands. These perceptions and knowledge are built over time, based on experiences. The product market outcome is based on the logic that the impact of the brand should be indicated by the company’s performance in the marketplace. This theory explains the price premium reflected in strong brands. Other measures of this performance include market share, relative price, and other, more complex measurement methodologies. The financial market outcome measures the financial value of the brand. Some measurements of this outcome can include the price of the asset when it is sold or through discounted cash flow analyses.

The product market outcome is the market modeled in this dissertation research. This analysis is intended to demonstrate that strong and/or growing brands will increase the performance of the company in the marketplace. This increase can be demonstrated through price premium and/or through increased sales volume. The customer mind-set is also represented in the data collected to measure the strength of the corporate brand.

The data collected by Tenet Partners to measure BrandPower is an important component of this research, not only because it quantifies measures of awareness and perception of corporate brands, but because it is also available on a large enough scale in terms of the history and number of companies tracked to provide generalizable results. This Brand Power data eliminates the subjectivity that can be found in many other models (i.e., Gregory, 2018). The quantitative measures do not force the modeler or interpreter to make value judgments based on brands that are derived from their own biases. The BrandPower, Familiarity and Favorability measures are the collective opinion of the impartial observers living in the marketplace.

Gaps in the Literature

Despite the tremendous volume of literature on the topic of branding, a noticeable gap exists in the literature. This gap represents the topic of corporate brand impact on sales/revenue performance. Such a gap is likely due to a lack of information available to measure and quantify corporate brand strength on a significant scale. The absence of quantifiable data can be addressed through Tenet Partner’s Corporate Branding Index. This study provides the data with which to fill this gap in the literature.

Hypotheses

The reviewed literature has influenced the development of hypotheses related to how the corporate brand can contribute to growth in sales/revenue-per-share. Answering these hypotheses help to clarify the relationship between the corporate brand and sales/revenue-per-share growth.

Four hypotheses guided this dissertation, along with several tests to prove or disprove them. The hypotheses are outlined below, with “H” representing “Hypothesis,” in sequential order.:

H1: Growth in BrandPower will result in growth in sales-/revenue-per-share change.

H2: Change in Familiarity and Favorability will have different, but positive, impacts on

sales/revenue-per-share growth.

H3: The different attributes that make up Favorability have a different, but positive,

impact on

sales/revenue-per-share.

H : Various conditions, such as industry affiliation, will positively alter the impact of 4

brand factors on sales/revenue-per-share growth.

BrandPower, as identified in H1, is the interaction between Familiarity and Favorability.

The other hypotheses test the components of BrandPower or other conditions, such as industry and sector affiliation. Taken together, these hypotheses should help to understand whether corporate brand has an impact on sales/revenue-per-share generation, and if so, how. This understanding will help communications executives across industries offer prescriptive guidance regarding the allocation of efforts and resources to articulate the corporate brand.

This dissertation offers a quantitative analysis that is supported by the BrandPower data and other empirical data collected, representing paid media investment, financial performance, and analyst ratings.

CHAPTER THREE:

METHODOLOGY

Purpose

The purpose of the research in this dissertation is to identify whether a linkage between the corporate brand and sales/revenue-per-share growth exists. The data utilized for this dissertation is the CoreBrand® Index supplied by Tenet Partners, the paid media investment obtained from Kantar Media Intelligence and the financial data gathered from Yahoo!’s “Finance and Value Line Investors Survey.” The research question being explored is: “How does the performance of the corporate brand impact sales/revenue-per-share?” This data has previously been examined to identify the impact of the corporate brand on stock performance (Gregory,

2018), but this study is the first time it is being used to examine the corporate brand’s impact on sales/revenue-per-share growth.

Sales/Revenue-per-Share Explained

Sales/revenue-per-share is the total sales of the company divided by the number of total shares of stock in the company. As a simplistic example, if a company had $100 in total revenue and issued 10 shares of stock to the public, its sales/revenue-per-share would be $10.

Sales/revenue-per-share is the dependent variable for the analytics in this dissertation.

The goal of this analysis is to predict how fluctuations in a company’s corporate brand impact a company’s sales/revenue-per-share over time. Specifically, the study explores whether a growing

19 corporate brand will result in that company seeing gains in its sales/revenue-per-share over the same period.

Practical Business Application of the Model

The use of the model for the practitioner is intended to help him/her understand the corporate brand’s contribution to sales-revenue growth. As executives allocate scarce resources to communicating their corporate brand and its messages, the executive needs tools in place to track the performance of his/her corporate brand and its impact on business results. This analysis is one piece of the brand-intelligence mosaic that should be used to help companies develop their communications plan. Companies should utilize data gathered at the strategic, operational, and tactical level of their organizations. Understanding the corporate brand’s contribution is but one piece at the strategic level.

Ultimately, it is hoped this work can change the way the corporate brand is perceived.

The sought-after outcome is that senior corporate leadership evolves its thinking relative to the value of the corporate brand. Over time and with the right evidence, it is desirable for senior leadership to view the corporate brand as an asset that has value and should be invested in, rather than being perceived as an expensive nuisance that consumes resources with dubious return.

Research Design and Data Collection

The following is the data collection methodology for the CoreBrand® Index (CBI):

BrandPower is a proprietary measure of corporate brand strength developed by Tenet Partners for its consulting practice. Based on research collected in the CoreBrand® Index Study, BrandPower is a measure that combines Familiarity and Favorability attributes into a measure of the size of a company’s audience and its perceptions of that company.

Respondents are impartial observers, consisting of senior executives within the largest companies in the United States. They represent senior business leaders and affluent consumers.

The companies have disposable income and investable assets (Gregory, 2018).

The CoreBrand® Index has been in use since 1990 to evaluate corporate-brand strength of companies, to build brand-valuation models, and to produce predictive return-on-investment models for corporate brands. No data set like this exists anywhere else. Approximately 1000 brands have been reported annually from 1990–2001; from 2002 to the present, the data has been reported quarterly.

The following factors are tracked:

• Familiarity – based on respondents who know more than just the company name.

• Favorability – a measure of perception of a company based on average of the below

attributes:

• Overall Reputation – a measure of the general perception of the

company.

• Perception of Management – a measure of the quality of the

management and leadership of a company.

• Investment Potential – a measure of willingness to invest in a

company.

• BrandPower (the interaction of Familiarity and Favorability).

• Culture of Innovation (an attribute that is tracked but is not a part of BrandPower; it

measures the degree to which innovation is in the DNA of a company).

This data was designed to be a quantitative representation of the corporate brand to be used as a variable in business modeling. BrandPower and its components are the centerpieces of

21 this research. The results of this study are intended to help managers understand the impact that the corporate brand has on business results and how to better allocate resources to direct it.

Tenet Partners secured a license to use this data in the present research (see Appendix A).

The only limitations on the license were that it be used for academic research and not for profit.

The most recent data was not to be used; a verbal agreement was made to use data up to fourth quarter 2016. Company names were to be sanitized in order to protect the salability of the data for Tenet Partners.

Respondents were sourced via a purchased telephone list. The specifications for the list were: V.P., Director, Senior Manager, and higher respondents in the top 20% of U.S. businesses, based on revenue in the United States. Only one respondent per any company could respond per quarter. Respondents were random and do not represent a panel. It is possible for a respondent to respond more than once over history, but repeat participants were not sought, so the incidence rate of repeat participants is extremely low.

I conducted 8,600 telephone interviews per year. Each company was rated by 100 respondents per quarter, totaling 400 responses each year. The survey contains a list of 40 companies rated for Familiarity and the three Favorability attributes. I tracked 860 companies and used fourth quarter data (which represents each full year). This research represents the 400 interviews conducted in each year.

The telephone interview was relatively simple. It was designed to result in data that provided a quantitative variable that represented a company’s corporate brand and key components. This data is intended to be used in the context of other corporate data, such as fundamental financial data and marketing spending.

Telephone interviewers used a system called “ConfirmIt.” This system displays a survey on the computer screen and randomly selects companies to be queried, ensuring that each company on the survey had the same number of respondents. The interviewer recorded the respondents’ results and then ConfirmIt warehoused the data. This system has been in use since

1998; prior to that, interviewers recorded results with pen and paper.

A limitation of the research was the audience was limited to a sub-set of the total population. While this limitation was intentional, a portion of the total population was excluded.

Also, responses were on a 100-point scale, meaning there is an upper and lower limit on the

BrandPower data while financial data is on a continuous scale and can always grow or decline.

Therefore, with BrandPower, it is important to understand the impact of the absolute level of

BrandPower as well as its rate of growth.

When interviewing, respondents were first read a brief introduction, explaining the purpose of the study and ensuring confidentiality. Finally, the importance of the respondents’ participation was explained, and they were thanked.

Next, the respondents were asked to rate their Familiarity with a list of 40 randomized companies. Familiarity was rated using a 5-point scale: 1=Unfamiliar, 2=Know the Name Only,

3=Somewhat Familiar, 4=Familiar, 5=Very Well Known.

Then, if respondents rated a company 3 or higher, indicating they knew more than just the name of the company, they were asked to rate their Favorability towards the company on three key attributes. The attributes were: Overall Reputation, Perception of Management, and

Investment Potential. The attributes were each evaluated on a 4-point scale: 1=Poor, 2=Fair,

3=Good, 4=Excellent.

The data was tabulated based on frequency distribution of the results. To calculate

Familiarity, the percentage of respondents to give each scaled response was calculated. Then, these results were weighted with a response of 3 being weighted as a 1, a response of 4 being weighted as a 2, and a response of 5 being weighted as a 3. These results were summed and then divided by 3, to reduce the score to a 100-point scale. For example, if Company A’s Familiarity results were as follows: “Unfamiliar” = 10%, “Know the Name Only” = 20%, “Somewhat

Familiar” = 40%, “Familiar” = 20%, and “Very Well Known” = 10%, the result would be:

(10 x 0 + 20 x 0 + 40 x 1 + 20 x 2 + 10 x 3) / 3 = 36.7 Familiarity Score. I refer to these as scores because they are weighted-average percentages.

Each of the Favorability attributes was calculated in a similar fashion. For example, if

Company A had Overall Reputation ratings of Poor 20%, Fair 10%, Good 60%, Excellent 10%, the result would be: (20 x 0 + 10 x 1 + 60 x 2 + 10 x 3) / 3 = 50.0 Overall Reputation Score.

The other Favorability attribute scores were tabulated the same way, and then scores for the three attributes were averaged together to create a Favorability score.

Then, BrandPower was created by multiplying Familiarity x Favorability x 0.01. For example, if Company A’s Familiarity score was 36.7 and its Favorability score was 50.0, its

BrandPower would be: 36.7 x 50.0 x 0.01 = 18.4 BrandPower score.

In 2016, a new attribute was added to my study: “Culture of Innovation.” It was designed to be a measure of the innovation in a company’s DNA; therefore, the measure is more than just product- or service innovation. This attribute was scored and tabulated the same way the other

Favorability attributes were, but it was independent and not incorporated into BrandPower.

Key Descriptive Statistics

The next step was identifying key descriptive statistics on the variables in the analysis.

Appendix A identifies several of the model variables, to offer a better understanding of the dimensions of the data. It also contains a key for all variable abbreviations in this dissertation.

The most important number on this table is the mean for the sales per share (SPS) 5-yr growth%. This is the sales/revenue-per-share, 5-yr % change. This number was used in the target

(dependent variable) in the chi-squared automatic interaction detector (CHAID) model. The companies studied were broken into high and low sales/revenue-per-share, 5-yr. % change companies. This model allows predictions of whether a company would be a high- or low- growth company, based on whether their company falls above or below the mean of 20.6%.

When evaluating the corporate brand, I found that, in Year 0, BrandPower ran from a low of 0.3 to a high of 82.5, with a mean of 29.3. For Familiarity, in Year 0, there was a low of 1.9, a high of 98.2, and a mean of 43.3. For Favorability, in Year 0, there was a low of 12.3, a high of

91.4, and a mean of 63.3. The Favorability attributes generally follow the same pattern as

Favorability. All scores are on a 100-point scale. This result means that BrandPower had a lower ceiling than Familiarity or Favorability. Familiarity ran almost the full range of the scale; meanwhile, Favorability had a higher floor, but a lower ceiling than Familiarity. This result indicates that a company can have almost any level of Familiarity, but respondents are not likely to rate a company zero or 100 for Favorability. Thus, BrandPower tends to be more sensitive to

Familiarity because it has a broader range of scores. For this reason, the BrandPower score-- which is the interaction between Familiarity and Favorability--is important, but understanding all its components is critical for a deeper understanding.

One weakness of BrandPower analysis is a crisis situation. Often, in a crisis, a company’s

BrandPower score increases. This increase does not make intuitive sense, but there is a logic to it. In a crisis, the company tends to be heavily covered by the media, resulting in increasing

Familiarity. However, the nature of that media coverage is often negative, resulting in decreased

Favorability. For that reason, it is important to understand the data to know what it tells us.

Another interesting finding in the data was a precipitous decline in ad spending from Year

-9 to Year -1 (no data was collected in Year 0). While the minimum for each year was 0.0, both the maximum and the mean over that period declined significantly. More research will need to be done to determine if it is an actual decline in marketing investments or a reallocation of marketing resources to other channels that are not considered advertising.

Key

The following abbreviations are used in the charts and tables; names are written out in the text:

Fam – Familiarity Rep – Overall Reputation Mgt – Perception of Management Inv – Investment Potential Fav – Favorability BP – BrandPower Innov – Culture of Innovation CF – Cash Flow EPS – Earnings per Share SPS – Sales/Revenue per Share DPS – Dividend per Share BVPS – Book Value per Share SO – Shares Outstanding Sls – Sales / Revenue Stk – Stock Price per Share

CHAID Model Overview

A chi-squared automatic interaction detector (CHAID) analyzes large data sets that are complex, creating non-binary decision trees. The decision trees can have an unlimited number of branches and are, therefore, wider than trees using binary methods. Initially. CHAID was developed for use in marketing models, where finding the best solution often involves more than one potential answer. CHAID algorithms are designed to identify the best predictive variable at each branch of the decision tree, which is done using F tests where the CHAID merges statistically similar (homogeneous), independent values with the dependent variable, and it maintains all other dissimilar (heterogeneous) independent variables (Gregory, 2018;

International Business Machines Corporation [IBM], 2013).

Exhaustive CHAID identifies the best split for each predictor-independent variable, based on adjusted p values. Because the dependent variable (sales/revenue per share 5-yr % growth) is continuous, exhaustive CHAID was used in the present analysis to provide the clearest answers to the question raised in this dissertation (Gregory, 2018).

Continuous predictive-independent variables were binned into a set of ranking categories.

This binning is the first step in the exhaustive CHAID process, and it is completed for each scaled predictor in the model. (Gregory, 2018; IBM 2013).

There are compelling reasons for utilizing CHAID instead of multiple regression analysis. Multiple regression analysis is often flawed due to multicollinearity while CHAID is not impacted by multicollinearity. CHAID also has greater explanatory power, which results in greater predictability in the model (Gregory, 2018).

Also, CHAID does not make assumptions of parameters based on frequency distribution.

It provides a high level of confidence that the best predictive variables are identified when

27 utilizing an evidence-based analytical process, which is standard in quantitative market research.

CHAID is the ideal analytics method for identifying and determining the relationship between independent and dependent variables (Gregory, 2018). For these reasons, CHAID is the best analytical process for this dissertation.

CHAPTER FOUR:

RESULTS OF THE STUDY

Analysis

The analysis began with data exploration. The non-modeling phase is designed to familiarize the researcher with the data and the relationships between the variables. This step reveals how the data interacts; it is a key step in understanding why model-interactions work, which is fundamental to building a model that does not only make predictions, but makes sense and has utility for practitioners and their businesses. This step is the foundation of the process

(Jank, 2011).

The data exploration phase is followed by the modeling phase. During this phase, the

CHAID model was employed, using the software to identify the relationships between the predictor variables and the target variables. In this case, the objective was to identify whether a company is expected to rise into the high sales/revenue-per-share category (> 20.6% growth) or fall into the low sales/revenue-per-share category (< 20.6% growth).

Data Preparation

The first step in the data preparation process was to merge the various data spreadsheets that composed the dataset for this analysis. The process included BrandPower, paid media (ad- spend), and fundamental financial data. This process resulted in a total of 845 companies in the dataset. However, further data prep was required.

The next step in the data preparation process was to further narrow the dataset in cases where BrandPower data was missing, which included companies that were dropped from the survey or were too recently added to be considered. Companies were included if they had brand measures available from Year 0 (the target year for the dependent variable) through Year -5, five years prior to the dependent variable.

Per the license to use the data, companies were assigned a random number and company names were removed to sanitize the companies (a key has been retained, so that any anomalies can be examined). Next, the data was sorted based on BrandPower in 2016; any companies that were not included in the survey in 2016 were removed. Then, the data was sorted based on

BrandPower in 2011; any companies that were not included in the survey in 2011 were removed.

No companies were covered in 2011 and 2016 that were not covered in the intervening years.

Then, the decimal positions were set based on the type of data. For example, BrandPower data has a single decimal while figures representing dollar values have two decimal places. Changes were calculated for all the data for 1, 3 and 5-year historical change. These changes were calculated based on percentage change as well as slope for all variables. Where available, projections for change in financial data were calculated for 1 and 3-year change. These projections also were calculated based on percentage change and slope of change.

With these data preparation steps completed, the total universe of companies left for analysis was 621 companies. At this point, the data was prepped for initial data-exploration, a step in which four more companies were removed, resulting in 617 companies modeled.

Initial Data Exploration – Univariate Test

The first step in the analytic process is data exploration. As defined in Jank’s (2011) textbook, data exploration is not a deep statistical analysis or modeling process. Rather, it is a

30 process in which the analyst familiarizes himself/herself with the data. This process allows the analyst to better understand the data so that s/he is better informed when it is time to construct a model. The first analysis performed in my study was a quintile analysis. The data was first sorted from highest to lowest, based on BrandPower in Year 0. Interesting findings are revealed in

Table 1 below.

Table 1. Quintile Analysis of Key Data for Companies in the Database.

BP Year 0 BP % 1-yr BP % 3-yr BP % 5-yr BP slope 1-yr BP slope 3-yr BP slope 5-yr Tier 1 65.1 0.888 3.690 6.600 0.552 0.765 0.712

Tier 2 45.6 2.140 8.990 17.588 0.872 1.091 1.057 Tier 3 22.7 5.011 20.729 45.299 0.882 0.788 0.568 Tier 4 9.8 10.563 27.139 20.453 0.657 0.291 -0.182 Tier 5 3.2 0.322 -1.929 -21.021 -0.099 -0.133 -0.376

Companies with the strongest corporate brands work to maintain their corporate brand strength because they have already grown them. Their sales/revenue growth is based on the strong corporate brands they have already built. By contrast, companies with the weakest corporate brands have yet to build corporate brands with enough leverage to impact sales/revenue growth. Those companies either continue to struggle, go out of business, are acquired, or grow their corporate brands and increase success. Companies with mid-level corporate brands reside in the “sweet spot;” their corporate brands are strong enough to be leveraged to grow sales/revenue-per-share, and they have room to grow them. These findings are consistent with findings in CoreBrand research conducted for BusinessWeek and CoreBrand/

Tenet’s Brand Laboratory research.

The shaded cells in Table 1 confirm findings from Koch et al. (2019) paper published by the American Society for Competitiveness, which asserts that the highest rates of growth for

BrandPower occur in the middle tiers.

The quintile analysis also showed a directional relationship between corporate brand and ad spending:

• Tier 1 - $278.1Mil.

• Tier 2 - $91.2Mil.

• Tier 3 - $50.5Mil.

• Tier 4 - $21.8Mil.

• Tier 5 - $11.5Mil.

This finding is expected to be key to uncovering return on investment (ROI) for communications spending and to help identify which companies should be poised to take the most advantage of communications spending.

Suttle (2020) may explain what is seen in the quintile analysis in Table 1 (above). Tier 1 brands are more mature and may not have room to grow. Tier 5 brands are small and need to achieve critical mass before they can grow. Middle tier brands have mass and awareness but have not totally matured, yet they do have opportunity to grow.

The shape of the relationship (non-linear) provided further justification that CHAID analysis would be the most appropriate methodology. The fact that growth rates are different at varying levels of corporate-brand strengths lends itself to the methodology’s binning of the variables.

Correlation Analysis

The next step in the data exploration stage was to conduct a correlation analysis between the independent and dependent variables. Because there were so many variables in the dataset, a traditional correlation matrix would have been unwieldy. Instead, the brand factors, BrandPower,

Familiarity, Favorability, Overall Reputation, Perception of Management, Investment Potential

32 and Culture of Innovation and their associated rates of growth were correlated to the 1- year, the

3- year and 5- year sales/revenue-per-share % growth rates to help identify where the relationships were strongest (see Appendix B).

The findings of the correlation analysis indicated that a stronger relationship existed between the brand factors and the sales/revenue-per-share % growth rate over 5- years than in 1 or 3- years. The longer the history of data, the stronger the relationships were. This finding is likely true for a finite period of time; the suspicion is that at a certain point, this result will diminish. This result suggests that brand consistency is an important component in driving sales growth over time.

The other finding of this analysis was that 1 and 3-year BrandPower growth were the most correlated with 5-year sales/revenue-per-share growth; also important was 1-year growth in

Investment Potential. All these showed a strong relationship. Other influential factors are

Familiarity level in Year -3, Year -2, Year -1, and Year 0; the 5- year Familiarity growth rate; the

1- year growth in Perception of Management and Investment Potential; and the 5-year growth in

BrandPower.

Again, these correlations demonstrate that the corporate brand takes time and consistency to translate into business growth. Consultants have presented this consistent theme to their clients for years, and it supports the concept that investment in communications should be consistent over time, rather than being erratic. Corporate communications should be seen as an investment rather than simply an expense.

The result of the correlation analysis also supports the hypotheses that suggest that

BrandPower and its components have a separate and varied impact on sales/revenue-per-share growth over time.

Hypothesis Testing

The present study performed various tests to confirm or reject the null hypotheses for this research. The objective of these hypotheses was to determine if various corporate-brand factors impact sales/revenue-per-share growth. In H1, H2, H3, and H4, the factors tested were BrandPower and its components, Familiarity and Favorability. Often, Familiarity is expressed as the awareness of the company and, in this case, it consists of a respondent who knows more than just the company name. Favorability is an average of the attributes Overall Reputation, Perception of

Management, and Investment Potential. These attributes typically represent a hierarchy of one’s perception of a company. Typically, Overall Reputation is the highest-rated attribute and represents an easier attribute to rate highly. It asks if a respondent has an overall positive perception of the company. Perception of Management is usually the second highest rated attribute and represents a deeper commitment, as it specifically addresses the leadership and the people in an organization. Typically, Investment Potential is the lowest rated attribute, for it represents a more difficult level of commitment; it is an indication of the respondent’s willingness to invest in the company. A fourth attribute that is not a part of the BrandPower calculation is Culture of Innovation; though this attribute was not a part of my hypothesis testing, it was examined. Culture of Innovation is a single attribute that seeks to evaluate the innovation climate within an organization. It is not a measure specifically of product, service, or process innovation; rather, it is a measure of the innovation inherent in a company’s DNA.

H4 tests the industry affiliation impact on a company’s sales/revenue-per-share growth. In the CHAID analysis, this hypothesis identifies how the impact of corporate brand differs for companies in various lines of business. Sector and industry affiliation were tested as part of the

CHAID analysis. In the CHAID analysis, the best way to test H1, H2, H3, and H4 is

34 simultaneously. This type of testing allows the model to evaluate the importance of each factor relative to one another. When the CHAID analysis evaluates all factors, it can identify the true impact of each factor on the dependent variable (IBM, 2013).

CHAID Analysis – Limited Micro Model

The key analytical process employed in this research is the CHAID analysis. This tool was used to build the analytical model. The model was applied to analyze the impact of corporate brand on sales/revenue-per-share growth at the macro and micro level. At the macro level, 617 companies were analyzed across sectors and industries. By contrast, at the micro level, sector and industry level data were analyzed in isolation to determine if effects vary based on business category. Unfortunately, while sector data could be modeled, the samples became too small to measure individual industries. Nonetheless, the macro model was able to reveal that industry affiliation is an important predictor.

The first analysis performed at the macro level was to examine industry affiliation and the brand factors impact on sales/revenue-per-share 5-yr % growth rate. First, the analysis identified which factors were the most important (Figure 1).

Figure 1. Impact of Industry and Corporate Brand Factors on 5-yr Sales / Revenue Per Share % Growth Rate.

Figure 1 demonstrates that when industry affiliation and corporate brand factors are considered in isolation from the financial factors, industry affiliation is the most important predictor of sales/revenue-per-share 5-yr % growth. This finding is understandable, as companies in different industries tend to grow at different rates.

After industry affiliation, Familiarity in Year -1 and Perception of Management in Year 0 were the most important predictors of sales/revenue-per- share 5-yr % growth, which is important because it means the level of performance of these variables matter. To grow sales, a company must be well- known to the public. If unknown, it is much harder to sell products. The

Perception of Management level is critical because it represents a level of trust between the company and the public. A high Perception of Management means the leadership is trusted; if management is not trusted, it creates a headwind that will impede sales.

The next most-important predictor is Investment Potential 3-yr slope, which is an indicator of absolute growth in Investment Potential, indicating that companies that inspire investment-confidence also grow their sales.

BrandPower 1-yr slope is the next most important predictor, which indicates that the absolute change in the summation of the factors that comprise the corporate brand is important for sales/-revenue-per-share growth. It indicates that companies need growth momentum in the corporate brand to grow sales.

Investment Potential level in Year -2 was also a predictor. This variable, when considered with Investment Potential 3-yr slope, indicates that not just Investment Potential growth, but also level are important factors. Companies need to be strong and regularly improve on this measure.

Again, this predictor is an indicator of investor confidence in the organization.

Familiarity 1-yr slope and Familiarity Year -5 were predictors of sales/revenue-per-share

5-yr % growth. These factors, in conjunction with Familiarity Year -1, clearly show that a company must be well-known and growing in that area to drive sales growth.

Figure 2. CHAID Decision Tree Macro Analysis, Industry and Corporate Brand Model.

Figure 2 shows the decision-tree model for the macro analysis that includes industry affiliation and the corporate-brand components’ impact on sales/revenue-per-share 5-yr growth.

The CHAID analysis examines factors that impact the dependent variable and looks for differences in impact of the independent variables, if those dependent variables have different impacts based on how they perform. The model looks for splits in the data and identifies all of them, as long as the P-values fall below 0.05.

The dependent variable, sales/revenue-per-share 5-yr % growth, was split into two bins: high growth was above 20.6% and low growth was below 20.6%. In this case, the top predictor was industry affiliation. Below, Table 2 identifies the industries contained in each of Nodes 1–4 in the decision tree.

Companies in Node 1 were influenced by Familiarity level Year -5, which indicates these companies were most impacted by Familiarity level at the start of the timeline.

Node 2 companies did not have other influencing variables, indicating that among the companies in that particular node, the corporate brand may be of lesser importance than in other industries. It is important to note that this node had the fewest industries in it.

Node 1 Node 2 Node 3 Node 4 Aerospace Commercial Banks Diversified Services Home Appliances Apparel, Shoes Computer Software Electronics, Electrical Equipment Hotel & Entertainment Beverages Computers & Peripherals Food Restaurants Building Materials Diversified Financial Paper Products Retailers Chemicals Furniture Telecommunications Industrial Equipment Auto Parts Metal Products Brokerage Metals Diversified Industrials Crude Oil Home Builders Motor Vehicles Insurance Petroleum Refining Internet Pharmaceuticals Medical Supplies & Services Publishing & Printing Office Equipment Rubber & Plastics Packaging Scient, Photo, Cntr Eq Educational Services Toiletries, Household Products Pharmacy Serv. Textiles Consulting Tobacco Distilled Spirits Transportation Athletic Equipment & Apparel Power & Energy Electric Utilities

Figure 3. Industries Included in Nodes 1-4 of the CHAID Analysis.

Node 3 had the most industries in it. These industries were most influenced by

Familiarity in Year -1. Companies with lower Familiarity were most influenced by Investment

Potential in Year -2 because the performance in Nodes 7 and 8 indicate that high Familiarity companies in Node 8 have a positive impact on performance, and low Familiarity companies in

Node 7 were lower performers. Familiarity Year -1 appears to have been negatively influenced by Investment Potential for these companies. Perception of Management 2016 influenced the higher performing companies in Node 8, indicating the importance of trust in corporate leadership.

The companies in Node 4 were overwhelmingly influenced by 1-yr BrandPower growth.

Node 9 shows that companies with low BrandPower growth also are projected to have low sales/revenue-per-share 5-yr growth. Companies in Node 10, which is high growth, were reactive to high BrandPower growth.

This model only accounts for the BrandPower components and industry affiliation. Due to its extreme importance, industry association was left in as a predictor. The application of this model is more for the communications team within an organization; its use could help in prioritizing messaging for any communications program. By comparing an individual company’s performance on the BrandPower components to predictor importance, communications programs and messaging could be adjusted.

CHAID Analysis – Full Macro Model

The next step in the research was evaluating all the independent variables and assessing their impact on sales/revenue-per-share 5-yr % growth rate. The key predictors are shown in in

Figure 3. Once again, the most important predictor was industry affiliation, indicating that within industries, companies have similar growth characteristics. The next most- important predictor was dividend per share 3-yr growth rate. This relationship indicates that companies paying a higher dividend are more likely to be growing companies. The next factor is BrandPower 3-yr slope, which indicated that BrandPower growth is an important predictor of growth in sales/revenue-per-share 5-yr % growth rate. Investment potential level in Year -4 was the next most-important predictor, which indicated that a company perceived as a worthwhile investment is a leading indicator of future sales/revenue-per-share growth. Earnings predictability, an analyst ranking, was the next most important predictor. It indicates that companies with more stable, predictable earnings will grow more than others. The final important predictor of sales/revenue-per-share 5-yr % growth was stock price 5-yr % growth, which indicated that sales/revenue-per-share growth and stock price growth are linked.

The next step in the analysis was to evaluate the CHAID decision tree in figure 4. Once again, the dependent variable at the top of the tree is sales/revenue-per-share 5-yr % growth. A total of 617 companies were evaluated in the analysis.

Figure 4. Predictor Importance in Full Macro Model.

Figure 5. CHAID Full Macro Model Decision Tree.

At the top of the decision tree in Node 0, one sees that of the 617 companies evaluated,

379 were predicted to have high 5-yr growth (above 20.6% growth), and 238 were predicted to have low 5-yr growth (below 20.6% growth).

The next level of the decision tree is Dividend-per-share 3-yr growth percentages. These were examined in Nodes 1–5. Nodes 1 and 2 show negative- and low-dividend growth rate (up to

21.7% growth), and the companies were predicted to have low sales/revenue-per-share 5-yr % growth rates. Node 3 shows companies with Dividend-per share growth rates between 21.7% and

65.6% growth. These companies showed a higher predicted sales/revenue-per-share 5-yr growth

40 rate. Of 138 companies in Node 3, only 71 companies were predicted to be high sales/revenue- per-share growth companies, and 67 were predicted to be low-growth companies. Node 4 companies are those above 65.6% Dividend-per-share growth. In this node, 66.6% of the companies were predicted to have high sales/revenue-per-share growth, and 33.3% of the companies were predicted to have low sales/revenue-per-share growth. Node 5 includes companies where no dividend was captured.

Nodes 6 – 16 show how industry affiliation impacts sales/revenue per share. Tables 3, 4, and 5 show how the industries break out into these nodes.

Node 6 Node 7 Node 8 Aerospace Diversified Services Brokerage Apparel, Shoes Electronics, Electrical Equipment Diversified Industrials Beverages Food Home Builders Building Materials Paper Products Insurance Chemicals Furniture Internet Commercial Banks Industrial Equipment Medical Supplies & Services Computer Software Metal Products Office Equipment Computers & Peripherals Metals Packaging Diversified Financial Crude Oil Educational Services Motor Vehicles Pharmacy Serv. Petroleum Refining Consulting Pharmaceuticals Distilled Spirits Publishing & Printing Athletic Equipment & Apparel Rubber & Plastics Scient, Photo, Cntr Eq Toiletries, Household Products Textiles Tobacco Transportation Power & Energy Electric Utilities Home Appliances Hotel & Entertainment Restaurants Retailers Semiconductors Telecommunications Auto Parts

Figure 6. Industries Included in Nodes 6, 7, and 8.

Node 9 Node 10 Node 11 Node 12 Node 13 Aerospace Computer Software Diversified Services Paper Products Home Appliances Apparel, Shoes Computers & Peripherals Electronics, Electrical Equipment Furniture Hotel & Entertainment Beverages Diversified Financial Food Industrial Equipment Restaurants Building Materials Metal Products Retailers Chemicals Metals Semiconductors Commercial Banks Crude Oil Telecommunications Motor Vehicles Auto Parts Petroleum Refining Brokerage Pharmaceuticals Diversified Industrials Publishing & Printing Home Builders Rubber & Plastics Insurance Scient, Photo, Cntr Eq Internet Toiletries, Household Products Medical Supplies & Services Textiles Office Equipment Tobacco Packaging Transportation Educational Services Power & Energy Pharmacy Serv. Electric Utilities Consulting Distilled Spirits Athletic Equipment & Apparel

Figure 7. Industries Included in Nodes 9, 10, 11, 12, and 13.

Node 14 Node 15 Node 16 Aerospace Diversified Services Home Appliances Apparel, Shoes Electronics, Electrical Equipment Hotel & Entertainment Beverages Food Restaurants Building Materials Paper Products Retailers Chemicals Furniture Semiconductors Commercial Banks Industrial Equipment Telecommunications Computer Software Metal Products Auto Parts Computers & Peripherals Metals Brokerage Diversified Financial Crude Oil Diversified Industrials Motor Vehicles Home Builders Petroleum Refining Insurance Pharmaceuticals Internet Publishing & Printing Medical Supplies & Services Rubber & Plastics Office Equipment Scient, Photo, Cntr Eq Packaging Toiletries, Household Products Educational Services Textiles Pharmacy Serv. Tobacco Consulting Transportation Distilled Spirits Power & Energy Athletic Equipment & Apparel Electric Utilities

Figure 8. Industries Included in Nodes 14, 15, and 16.

Macro Model Results

Once the predictors and their interactions had been identified, the results of the model had to be assessed. Figure 5 identifies the success of the model.

Figure 9. Results of the Macro Model.

This model performed with an 85.9% accuracy rate, which means that, in nearly 86% of its estimate, the model was correct. Of 617 predictions, 530 were correct while 87 were incorrect. Of 379 high predictions, the model predicted correctly 337 times, the sales/revenue- per-share 5-year growth to be above 20.6%. It predicted sales / revenue per share 5-year growth to be below 20.6% correctly 193 times out of 238.

This accuracy rate is high and could likely be improved over time as more is learned about the relationship between the corporate brand and sales/revenue-per-share growth. Also, as more data becomes available, the model can be improved.

Additional Models

Once the macro model had been constructed, additional work was needed at the sector level. The sectors under consideration were consumer staples, industrials, and technology. These sectors were considered because they are vastly different in terms of product, customer, supply chain, and production. If there were differences in the way the model performed, based on sector affiliation, it could be proven because these sectors would reveal the diversity.

For each sector, the same data-set was utilized, and the same process was followed. The

CHAID analysis identified the predictors of sales/revenue-per-share 5-year growth, by dividing them into high- and low 5-year growth bins (+/- 20.6%). Next, a decision tree was developed to identify the relationships. Finally, I assessed the results.

Consumer Staples Sector

I tracked fifty companies in the consumer staples sector. These companies produce everyday household items that day-to-day shoppers buy everywhere, from local shopping malls to grocery stores. These products are purchased by consumers from the top to the bottom of the economy.

The results of the model in the consumer staples sector were somewhat surprising. Figure

6 indicates that two predictors were of importance in the data set. The most important predictor was stock price 5-year % growth, which indicated that companies’ sales are growing and increasing their share price. The next most important predictor was the BrandPower 1-year slope, which indicated that growth in the corporate brand is critical in this sector.

Figure 10. Predictor Importance Consumer Staples Sector.

The CHAID decision tree in this study (Figure A1) shows that half of the companies were predicted to be high and half were predicted to be low sales / revenue per share 5-year % growth companies. Interestingly, companies without stock values, indicating they are private companies, were predicted to outperform the public companies.

Overall, this model received an 86% accuracy rate. As indicated in Figure 7 (below), 43 out of 50 predictions were accurate, and 7 were incorrect. High-growth companies were identified correctly 18 times, and falsely identified 7 times. Low-growth companies were predicted 25 times, all of which were correct.

Figure 11. Results of Consumer Staples Sector Model.

Industrials Sector

The next model for consideration is the industrials sector model, which represents everything from hand tools to aerospace. Some products are marketed to individual customers, but this sector is primarily business-to-business focused and sells large, expensive components or

44 machines. In this sector, 121 companies were represented. Once again, the same data set was considered, and I followed the same process as was developed in the macro model.

This sector has four important predictors of sales / revenue per share 5-year % growth

(Figure 8). The first is stock price 5-year % growth, indicating that companies’ stock growth is a predictor of sales / revenue per share growth. The next predictor is Familiarity 5-year slope, indicating that companies that are growing public awareness are more likely to create sales / revenue growth. The next most important predictor of sales / revenue per share growth is

Perception of Management Year -2, indicating that the level of esteem shown to the leadership of the company in recent years influences sales / revenue per share growth. Finally, projected 1- year future dividends paid out is also an important predictor, which is indicative that the company’s prospects are an important indicator of performance; this indicator is not surprising given the products sold are often large equipment and machinery.

Figure 12. Predictor Importance Industrials Sector.

The CHAID decision tree (Figure A2) shows that of the 121 companies, 55 are predicted to be high sales / revenue per share companies while 65 are predicted to be low-growth companies. The stronger stock-growth companies are predicted to outperform the lower stock- growth companies; however, the private companies are projected to do better than the publicly traded companies. High Familiarity companies lead to stronger Perception of Management, which is important to growth for private companies.

Companies in the industrial sector are most reliant on the corporate brand factors of

Familiarity growth and strength of Perception of Management, which means they need to be growing the size of their audience and demonstrating strong leadership. While each company needs to be evaluated on an individual basis, companies should consider some general advice.

First, a communications audit of the creative messaging and share-of-voice analysis would allow companies to evaluate their messaging and determine whether their communications spending in the category is adequate to develop a strong position relative to peers. Findings may indicate a need to strengthen messages relating to corporate leadership and vision. Also, the company might increase investment in the corporate brand to boost the Familiarity factor.

Figure 9 (below) demonstrates the accuracy of the industrials sector model. In this case, it was 81.8% accurate, with 99 correct predictions and 22 incorrect predictions. Meanwhile, of 56 high-growth predictions, 39 were accurate while 17 were incorrect. Of 65 low-growth predictions, 60 were correct and 5 were incorrect.

Figure 13. Results of Industrial Sector Model.

Technology Sector

The technology sector consists of companies from semiconductors, computers, and peripherals to cloud-computing companies, software, and internet companies. Fifty companies were considered in this sector. They were evaluated using the same data set and process used in

46 the macro model and the other sectors. The technology sector sells to individual consumers and business-to-business.

Four important predictors are in the technology sector (Figure 10). The most important predictor is dividend per share 3-year % growth. These are often cash-wealthy companies and may be an overall indicator of corporate health. The next most important predictor is industry affiliation within the sector, which is likely an indicator that the sector is very diverse, with some companies providing products and others providing services. Perception of Management in Year

0, the target year for sales / revenue 5-year % growth, showed that the absolute strength of leadership is an important factor. This result is likely due to the importance of securing personal and financial information for which these companies must be responsible. Finally, the 1-year slope in Favorability is the last important predictor of sales / revenue per share growth.

Figure 14. Predictor Importance Technology Sector.

In the technology sector, the most important corporate-brand factors are Perception of

Management strength and Favorability growth, which indicates that quality of image is most important in this sector. A company would want to ensure that its spending level is competitive with peers; even more important is the company’s vision must be clearly articulated and the general perception of the company must be on the rise. These issues can be addressed by communications’ messaging efforts.

The CHAID decision tree (Figure A3) indicates that companies in Node 3, computers and peripherals, were impacted by Familiarity growth over the previous year. By contrast, companies in Node 5, internet companies, were most influenced by Perception of Management in Year 0.

All high sales / revenue per share companies were among companies that were private or paying a higher dividend.

Figure 11 illustrates the results of the technology-sector model. Of 50 companies, 86% of the predictions were accurate. Specifically, 47 predictions were accurate and 7 were incorrect.

High sales / revenue growth was accurately predicted 29 times and inaccurately predicted 6 times. Low sales / revenue per share growth was predicted accurately 14 times and inaccurately predicted only 1 time.

Figure 15. Results of Technology Sector Model.

Comparing the Sector Models

Of the three models in this study, the consumer staples model was the simplest. It appears that this result reflects the nature of the low-complexity purchase decisions in that particular market, as compared to the others. Consumers in this market are looking for financial strength in the company and for growing brands.

A second model, industrial and technology, proved to be more complex, relying on more factors for sales / revenue per share growth. While the industrial sector was more reliant on long- and short-term corporate brand factors, the technology sector relied more on short-term brand

48 factors. This finding is intuitively understandable as industrial companies are most likely to have corporate brand reputations built over a long period of time and performance while in the technology space, new corporate brands emerge all the time.

These three sectors demonstrate the diversity of businesses and how the corporate brand works to support them and impacts them in different ways. This study also shows that those with responsibility for a company’s corporate brand must understand the bigger picture of what drives revenue for their company and industry.

Industry Analysis

While I attempted a more granular analysis of individual industries, using one industry sub-group from each of the sectors, the industry sub-groups proved to be too small in the number of companies to build a reliable model.

CHAPTER FIVE:

SUMMARY AND DISCUSSION

The concepts expressed in this dissertation are grounded in decades of corporate-brand research. The methods offered are non-traditional, non-GAAP, and may be debated by accountants and finance executives. However, the results are evidence-based and data-driven; they are not opinion. In fact, they are borne out in quantitative data.

As it relates to the concept of intangible capital, this research on the corporate brand’s impact on sales / revenue per share growth uncovers another layer of intangible capital. As stated in Ocean Tomo LLC’s study “Intangible Asset Market Value Study” (2017), 84% of a market valuation is based on intangible assets. This percentage is up 17%, compared to the first study of

1975. By 2020, the updated study indicates that the market valuation’s number increased to 90%.

This study was not based on market cap but on sales / revenue per share growth--which is yet another component of intangible value. The point is the corporate brand contributes to value creation through multiple avenues. Too often, brand consultants and advocates argue about the right valuation methodology or the best methodology. The truth is that there are multiple methodologies, multiple ways in which value is created and, in many cases, they are complimentary. The method put forth in this dissertation is a unique, new way to capture value creation. In the study of intangible capital, exploring brand and corporate-brand valuation is like peeling back the layers of an onion; the present study adds one more layer to our knowledge.

Still, multiple layers of brand-valuation have yet to be identified and explored.

Summary of the Problem

Consultants often discuss the value created by product brands and corporate brands.

However, historically, there has been little evidence. Subjective models based on experts in the filed have done little to boost credibility in the category. To tie the claims to the reality, hard data was needed, which is where Tenet Partners’ CoreBrand Index database is beneficial. This database creates quantitative variables, representing corporate brands and their various dimensions over time, in enough volume to be used as a variable in quantitative modeling. With such data, consultants can offer advice to their clients based on factual data and real analysis.

Rather than offering clients their opinions, consultants can offer factual data.

In my dissertation, the data offers answers to whether the corporate brand contributes to sales / revenue per share growth. This finding is important because it allows corporate executives the opportunity to choose between alternatives when allocating scarce budget funds based on expected returns, rather than relying on intuition and speculations. The executives’ allocation of resources can be viewed as an investment rather than merely a random distribution of funds.

Research Question

RQ: How does the performance of the corporate brand impact sales / revenue per share?

Hypotheses

H1: Growth in BrandPower will result in growth in sales revenue change.

H2: Change in Familiarity and Favorability will have different, but positive, impacts on

sales/revenue per share growth.

H3: The different attributes that make up Favorability have different, but positive, impact on sales/revenue per share.

H : Various conditions, such as industry affiliation, will positively alter the impact of 4 brand factors on sales/revenue per share growth.

Summary of the Methodology

The key to removing subjective bias from value-creation research is consistent quantitative research. The CoreBrand Index is the vehicle for creating this level playing field and removing all bias. For nearly three decades, the Index has utilized the same collection methodology and series of question to measure the health of corporate brands. The CoreBrand

Index is at the center of my research and provides the year-over-year consistency to allow comparisons of value creation across industries, companies, and time.

The importance of having quantitative data like the CoreBrand Index is it provides a distinct measure of the corporate brand for each company, over time. Many methodologies, such as discounted cash flow, make assumptions about brand strength based on the performance of the company. However, the methodology in this study allows the researcher to observe the brand as an independent variable. For example, it is possible for a company to have a strong brand yet underperform financially and vice versa. This approach allows for a more accurate, realistic observation of what is occurring in the marketplace. Undoubtably, more accurate observations and data lead to more accurate, precise models.

Summary of the Results

The research question in this dissertation was: “How does the performance of the corporate brand impact sales / revenue per share?”

In response to this question, sales / revenue per share 5-year growth was examined as the target (dependent) variable. BrandPower and all of its components (Familiarity, Favorability,

Overall Reputation, Perception of Management, Investment Potential, and Culture of Innovation)

52 were shown to have various impacts on sales / revenue per share 5-year growth, based on the results of the macro model and sector sub-models. These models accurately predicted sales / revenue per share 5-year % growth rates as high (above 25.6%) or low (below 25.6%) with a high degree of success.

• Macro model – 85.90% accuracy

• Consumer staples sector model – 86.00% accuracy

• Industrials sector model – 81.82% accuracy

• Technology sector model – 86.00% accuracy

The fact that corporate brand and/or its components had a positive influence in each of these models and improved their predictive powers supports the null hypotheses for H1, H2, and

H3. The fact that industry affiliation was the most important predictor in the macro model and that the impact of brand factors was different in the sector models supports the null hypotheses in

H4.

The present study supports the idea that the corporate brand can be measured as a quantitative variable. Thus, the contribution to sales / revenue per share growth can be measured as well.

Contribution to Knowledge

The present study is the first to measure corporate-brand contribution to sales / revenue per share growth on this scale. It is a foundational work in the area of brand valuation; it also contributes to the other measures of brand value, such as product/brand contribution to sales, corporate brand contribution to stock performance, royalty relief, discounted cash flow analysis, and many more measures. This study sheds light on what was already assumed-- the corporate brand could increase product sales--but had never been quantitatively proven.

Further, this study demonstrates a continued utility of the Corporate Branding Index, which has been shown widely to be a predictor of future value. The Index dataset has been used by Tenet Partners/CoreBrand in the development of return-on-communications-investment models in brand-valuation models (Gregory, 2018). This body of work shows the importance of building a quantitative database of this order.

It is expected that this work will further the knowledge of the value of intangible assets and their contribution to corporate performance. Hopefully, this study can serve to educate business executives and academics, to help implement these findings and expand the base of knowledge on the subject.

Practitioner Contribution

The intended primary contribution of this research is to help practitioners break the traditional mindset that product brands contribute to sales / revenue and corporate brands contribute to stock performance. That belief remains true, but the intention of this research is to illuminate the fact that corporate brands can also contribute to sales / revenue. Simply stated, when the company is well-known and respected, it is easier to sell products and services. The evidence herein should provide a foundation for further work to examine and quantify this contribution.

More systems like the CoreBrand Index need to be developed to give practitioners additional options to evaluate and master their brands. The key to the Index is it is consistent, simple, and makes intuitive sense. Since CEOs do not have the time to digest massive and complex measurement systems, systems need to be developed to help communications executives delve into the tactical components of their programs. The CoreBrand Index is

54 designed to be a strategic tool to help senior leadership understand the big picture. Such a tool spurs discussion, identifies strengths and weaknesses, and helps senior leaders make decisions.

This dissertation is intended to help business leaders and communications executives understand the importance of, and the contribution of, a strong corporate brand. Communications executives may consider this work as further evidence of the impact of a strong corporate brand and use it to justify expenditures on communications budgets to grow their brand. Senior executives may find this work helps them understand the reason to allocate budgets for supporting the corporate brand. Such budgeting is justified by the evidence herein.

Besides the CoreBrand Index, the present work builds on other research on corporate- brand value creation. I view my research as filling a gap in the existing literature. My study does not provide a perfect solution, but it is a first step into this wisdom of value creation. The literature on corporate-brand impact, as regards revenue generation, is small, largely because the data has not been made available on a large scale for an in-depth quantitative study.

Intangible capital, which is placing value on elements that have worth but traditionally are unrecognized, should be the guiding force for brand measurement and valuation systems moving forward. This concept is likely the most valuable one in this study, and the model herein is an important first-step to understanding corporate-brand’s contribution to revenue generation.

However, it must be viewed as one element of the corporate brand’s impact on businesses; further study and development of this concept will require many minds and perspectives.

However, understanding how brands fit within intangible capital is the key to understanding, unlocking, and unleashing the power of brands.

Study Limitations and Future Research

While this dissertation is a great first step and foundation into the area of corporate-brand contribution to sales / revenue per share growth, more work needs to be done.

Some recommendations for further study arose from my research based on the facts that

1) This study used only a single point-in-time observation for each company analyzed. 2) The data used covers the period 2011–2016, so the dependent variable is only sales / revenue per share 5-year growth in 2016. Further work should be done to improve the rigor and generalizability of the study by examining other points in time (for example, 2012–2017, 2013–

2018 and so forth). Doing so will increase the validity of the study and make the results more reliable for business leaders. Since product/brand data was not available to be examined as part of this study, I look forward to future research that includes product brands from major lines of business within these same companies to further specify which result is driven by the product brands and which is driven by the corporate brand. Future research is also needed to assimilate and synthesize the various contributions of product- and corporate brand to understand how the various components of brand work together to create value. This research would help communications executives better understand how to optimize their efforts to create brand valuation most efficiently.

The knowledge gained in this study should be used to create a forecasting model to understand how investments drive the brand and the corresponding value created. This knowledge will make the model more actionable in practice. The concept of brand-valuation modeling has likely uncovered only a small portion of the total value created for a company. It continues to be a fertile landscape for further research and exploration.

Conclusions

This study demonstrates not only how to measure the contribution of the corporate brand to sales / revenue per share growth, but it is an empirical approach to brand valuation. The objective has been to quantitatively measure the corporate brand and seek one of the dimensions of how it impacts a company, which is the basis of subjective, quantitative brand evaluation.

My analysis demonstrates that a macro view can be generalized by asking, “How does the corporate brand impact sales / revenue per share.” The key for a communications executive is to understand how the model behaves for his/her individual company. The model clearly demonstrates that industry affiliation is a key consideration for every company, which makes rational sense because brands impact various types of companies differently. For example, for one type of company, Familiarity may be the most important factor; but in others, Perception of

Management may be the more important factor. The sector models were based on vastly different types of companies and showed vastly different impact of the various brand components. These effects should be studied in more detail. In an ideal practice situation, these models would be customized to the extent possible for individual companies, based on their competitors and markets. As with most models, the decision on how granular to proceed should be weighed by the cost of the effort and the additional variance in the dependent variable explained.

The fact that the strength of the corporate brand can be quantitatively measured is not new. What is novel is that a brand’s impact on sales / revenue per share also can be quantitatively measured. More and more companies are embracing the importance of intangible capital. While this factor is not GAAP compliant, it can be used as a management tool to effectively manage the corporate brand and the optimization of intangible capital in value

57 creation. The model I offer here, like any other model, can be implemented and refined over time, as knowledge is gained. The most effective companies will use such tools rather than rely on experience, instinct, and/or intuition.

The objective in practice is not to secure unlimited funds for communications executives to flood the various channels of communications, but to understand the impact created from an informed investment decision to optimize the investment and return of brand valuation. A methodical approach should be used to measure brand valuation and impact. The data must be explored and understood before any modeling tools can be used with success. Like many other disciplines, brand intelligence can best be viewed as a mosaic. No one piece of information tells the entire story. The most successful companies will start by gathering and examining their data.

The relationships between the corporate brand components and other financial components will tell a compelling story that must be understood. The relationships between the communications inputs and the corporate brand components show us what can be done to articulate the corporate brand. Once these factors are understood, the model can be implemented to the greatest success.

Then, the communications input factors can be manipulated to have the greatest impact on the corporate brand to maximize its impact on business results, such as sales / revenue per share.

Once that is done, rates of return-on-investment can be considered and used to identify the point where return-in-business-results exceeds the effort and capital invested.

Without implementing a system like the one defined in this dissertation, companies are operating without sufficient knowledge, which is unacceptable. Companies must operate in an environment of data and modeling tools to assist decision making. Note that these tools assist decision making. Executives are still responsible for making final decisions, yet the tools need to

58 be utilized for the best decisions (and they ensure accountability). Decisions and data are key to successful business outcomes in today’s business environment.

REFERENCES

Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New York: Toronto: New York: Free Press ; Maxwell Macmillan Canada; Maxwell Macmillan International.

Aaker, D. A. (1996). Building strong brands. New York: Free Press.

Aaker, D. A. (2004). Leveraging the Corporate Brand. California Management Review (3), 6.

Abratt, R., & Bick, G. (2003). Valuing brands and brand equity: Methods and processes. Journal of Applied Management and Entrepreneurship, 8(1), 21-39. https://search.proquest.com/docview/203904398?accountid=14745

Ahmad, A., & Thyagaraj, K. S. (2014). Brand personality and brand equity research: Past developments and future directions. IUP Journal of Brand Management, 11(3), 19-56. https://search.proquest.com/docview/1628369270?accountid=14745

Ailawadi, K. L., Lehmann, D. R., & Neslin, S. A. (2003). Revenue premium as an outcome measure of brand equity. Journal of Marketing, 67, 1-27.

Balmer, J. M. T. (2001). Corporate identity, corporate branding and corporate marketing: Seeing through the fog. European Journal of Marketing, 35(3/4), 248-291.

Chandler, J., & Owen, M. (2002). Developing brands with qualitative market research. London, UK: Sage Publications, Ltd.

Clancy, K., & Krieg, P. (2007). Your gut is still not smarter than your head: How disciplined, fact-based marketing can drive extraordinary growth and profits. John Wiley & Sons, Incorporated. ProQuest Ebook Central. http://ebookcentral.proquest.com/lib/usf/detail.action?docID=290300.

Fischer, M. (2007). Valuing brand assets: A cost-effective and easy-to-implement measurement approach. http://www.marketing.uni-koeln.de/ fileadmin/wiso_fak/marketing_m/pdf/Fischer/Downloads/ Downloads_final_Abstracts/Fischer__2007__Valuing_Brand_Assets_Abstract.pdf

Gazley, A., Sinha, A., & Rod, M. (2016). Toward a theory of marketing law transgressions. Journal of Business Research, 69, 476-483.

Graydon, B. (2007). Product branding in howells's a modern instance. ANQ: A Quarterly Journal of Short Articles, Notes and Reviews, 20(2), 35-39. doi:10.3200/ ANQQ.20.2.35-39

Gregory, J. R. (1997). The impact of advertising on stock performance. New York, NY: American Association of Advertising Agencies.

Gregory, J. R. (2001). The bottom-line impact of corporate brand investment: An analytical perspective on the drivers of ROI of corporate brand communications. Journal of Brand Management, 8(6), 405-416. http://dx.doi.org/10.1057/palgrave.bm.2540040

Gregory, J. R. (2018). Intangible capital: Culture of innovation and its impact on the cash flow multiple [Doctoral dissertation, University of South Florida]. Scholar Commons.

Gregory, J. R., & Wiechmann, J. G. (1997). Leveraging the corporate brand. Lincolnwood, Il: NTC Business Books.

Gregory, J. R., Wiechmann, J.G. (1991). Marketing corporate image: The company as your number one product. Lincolnwood, Illinois USA: NTC Publishing Group.

Gujarati, D. N. (1995). Basic econometrics. New York, NY USA: McGraw-Hill, Inc.

Haigh, D. (2017). Understanding brand value differences. Brand Finance.

Hassan, H. B. (2014). Corporate brand extension value and the mediating effect of experience and brand personality (Publication No. 10188501) [Doctoral dissertation]. ProQuest Dissertations & Theses A&I; ProQuest Dissertations & Theses Global.

Holland, T. (2017). What is branding? A brief history. Sky Word. https://www.skyword.com/contentstandard/branding-brief-history/

Hui-Ming, D. (2010). Corporate social performance and financial-based brand equity. The Journal of Product and Brand Management, 19(5), 335-345. http://dx.doi.org/10.1108/10610421011068577

Hutton, K. (2017). Non-GAAP metrics are a response to faulty reporting standards [“Square-Off”]. Corporate Finance News and Events. http://ww2.cfo.com/regulation/2017/05/nongaap- metrics-response-faulty-reporting

International Business Machines Corporation. (2013). CHAID algorithms. IBM SPSS modeler 16 algorithms guide, 99-111.

Jank, W. (2011). Business analytics for managers. New York, NY USA: Springer Science+Business Media, LLC.

Keller, K. L. (2001). Building customer-based brand equity: A blueprint for creating strong brands. Report no. 01-107. Marketing Science Institute.

Koch, C., Puckey, B., & Williams, V. (2019). Empirical findings: The corporate brand. Competition Forum, 17,(1), 1-12.

Lagrost, C., Martin, D., Dubois, C., & Quazzotti, S. (2010). Intellectual property valuation: How to approach the selection of an appropriate valuation method. Journal of Intellectual Capital, 11(4), 481-503. http://dx.doi.org/10.1108/14691931011085641

Leitch, S., & Richardson, N. (2003). Corporate branding in the new economy. European Journal of Marketing, 37(7/8), 1065-79.

Liu, Y., Foscht, T., Eisingerich, T., & Tsai, H. T. (2017). Strategic management of products and brand extensions: Extending corporate brands in B2B vs. B2C markets. Industrial Marketing Management, 71, 147-159. https://doi.org/10.1016/j.indmarman.2017.12.016

Ludicke, M. K. (2006). A theory of marketing: Outline of a social systems perspective [Doctoral dissertation, University of St. Gallen].

Melewar, T., Gotsi, M., & Andriopoulos, C. (2012). Shaping the research agenda for corporate branding: Avenues for future research. European Journal of Marketing, 46(5), 600-608.

Mintz, O., & Curim, I. S. (2003). What drives managerial use of marketing and financial metrics and does metric use affect performance of marketing-mix activities. Journal of Marketing, 77(2), 17-40.

Ocean Tomo LLC. (2017). Intangible asset market value study. Ocean TOMO Intellectual Capital Equity. http://www.oceantomo.com/intangible-asset-market-value-study/

Peng, M. Y., & Li, Y. (2019). The impact of expressive brand relationship on brand loyalty: Drawing on relationship marketing theory. 2019 16th International Conference on Service Systems and Service Management (ICSSSM). doi:10.1109/ICSSSM.2019.8887611.

Puckey, B. (2012). Brand revenue: Measuring the brand’s ability to impact revenue. Association of National Advisors. http://www.ana.net/miccontent/show/id/ib-brand-revenue- corebrand

Rotfeld, H. J. (2014). The pragmatic importance of theory for marketing practice. The Journal of Consumer Marketing, 31(4), 322-327. doi: http://dx.doi.org/10.1108/JCM-02-2014-0854

Round, D. J., & Roper, S. (2012). Exploring brand name equity: Gaining insight through the investigation of response to name change. European Journal of Marketing, 46 (7/8), 938- 951.

Schultz, D. (2016). Market brand equity: Lost in terminology and techniques? The Journal of Product and Brand Management, 25(6), 507-515. https://search.proquest.com/docview/1826809442?accountid=14745

Slimane, K. B., Chaney, D., Humphreys, A., & Leca, B. (2019). Bringing institutional theory to marketing: Taking stock and future research directions. Journal of Business Research, 105, 389-394.

Suttle, R. (2020). Brands & product life cycle. Hearst Newspapers, LLC. https://smallbusiness.chron.com/brands-product-life-cycle-785.html

Xie, H. Y., & Boggs, D. J. (2006). Corporate branding versus product branding in emerging markets: A conceptual framework. Marketing Intelligence & Planning, 24(4), 347-364. http://dx.doi.org/10.1108/02634500610672099

APPENDIX A:

DESCRIPTIVE STATISTICS

Table A1. List of Variables and Descriptive Statistics.*

Field Std. Dev Min Mean Median Max Fam Y -5 30.3 2.4 41.7 33.7 96.8 Fam Y -4 30.7 2.1 41.6 33.1 96.7 Fam Y -3 31.2 2.6 41.6 32.8 96.9 Fam Y -2 31.6 2.1 41.9 32.0 97.5 Fam Y -1 32.0 2.3 42.5 31.7 98.0 Fam Y 0 32.3 1.9 43.3 32.1 98.2 Fam % 1-yr 11.4 -24.3 2.3 1.4 105.4 Fam % 3-yr 25.9 -51.1 5.4 3.4 205.3 Fam % 5-yr 52.7 -70.5 6.4 3.6 839.5 Fam slope 1-yr 1.8 -6.2 0.8 0.7 9.9 Fam slope 3-yr 1.6 -6.9 0.5 0.5 5.0 Fam slope 5-yr 1.6 -9.0 0.3 0.3 6.1 Rep Y -5 10.9 29.4 66.1 66.6 91.0 Rep Y -4 11.5 28.1 65.4 65.7 89.4 Rep Y -3 12.6 23.2 64.8 64.9 89.7 Rep Y -2 12.9 17.7 63.7 63.9 91.4 Rep Y -1 12.5 14.7 66.5 67.1 91.5 Rep Y 0 12.5 11.4 66.6 67.3 92.7 Rep % 1-yr 5.9 -30.6 0.2 -0.3 50.0 Rep % 3-yr 12.3 -57.8 3.4 1.6 74.8 Rep % 5-yr 15.2 -74.4 1.2 0.5 76.8 Rep slope 1-yr 2.9 -11.8 0.0 -0.2 12.5 Rep slope 3-yr 2.0 -7.3 0.8 0.6 7.3 Rep slope 5-yr 1.8 -7.9 0.1 0.1 6.0 Mgt Y -5 11.0 12.7 60.6 62.0 84.8 Mgt Y -4 11.7 19.6 60.4 62.0 86.2 Mgt Y -3 12.9 18.8 60.9 62.3 88.1 Mgt Y -2 12.6 15.9 62.0 63.0 89.1 Mgt Y -1 12.5 16.4 62.2 63.4 90.1 Mgt Y 0 12.2 12.9 62.6 63.6 91.5 Mgt % 1-yr 7.2 -24.9 1.0 0.2 66.4 Mgt % 3-yr 14.1 -50.7 4.0 1.2 121.7 Mgt % 5-yr 20.6 -70.2 4.5 2.9 282.8 Mgt slope 1-yr 2.9 -12.5 0.4 0.1 13.1 Mgt slope 3-yr 2.0 -7.8 0.5 0.2 8.4 Mgt slope 5-yr 1.7 -6.4 0.5 0.4 7.1 Inv Y -5 12.8 10.9 56.3 57.0 86.4 Inv Y -4 13.6 16.0 55.2 57.0 84.2 Inv Y -3 15.0 12.6 54.4 56.8 85.4 Inv Y -2 15.2 9.8 55.1 57.7 87.8 Inv Y -1 15.2 9.4 55.7 58.3 89.1 Inv Y 0 14.9 10.3 56.7 58.8 90.8 Inv % 1-yr 9.1 -26.4 2.5 0.9 55.2 Inv % 3-yr 20.6 -53.4 6.2 2.5 128.3 Inv % 5-yr 27.2 -68.2 2.4 1.1 323.9 Inv slope 1-yr 3.1 -12.4 1.0 0.5 12.8 Inv slope 3-yr 2.5 -7.5 0.8 0.5 9.8 Inv slope 5-yr 2.1 -8.1 0.1 0.2 6.9

Table A1 (Continued)

Field Std. Dev Min Mean Median Max Fav Y -5 11.2 24.6 61.0 61.6 86.4 Fav Y -4 11.9 23.3 60.3 61.3 84.5 Fav Y -3 13.1 19.7 60.1 61.5 86.9 Fav Y -2 13.1 15.3 60.2 61.4 89.0 Fav Y -1 13.0 15.8 61.5 63.1 90.1 Fav Y 0 12.8 12.3 62.0 63.3 91.4 Fav % 1-yr 6.5 -25.1 1.1 0.3 54.3 Fav % 3-yr 13.8 -51.9 4.2 1.9 96.7 Fav % 5-yr 17.6 -70.9 2.3 1.6 102.2 Fav slope 1-yr 2.7 -11.2 0.5 0.2 11.3 Fav slope 3-yr 2.0 -6.9 0.7 0.4 8.0 Fav slope 5-yr 1.8 -6.9 0.2 0.3 6.0 BP Y -5 22.2 1.0 27.6 20.9 83.4 BP Y -4 22.3 0.8 27.5 20.8 81.7 BP Y -3 22.9 0.6 27.7 20.6 81.0 BP Y -2 23.0 0.4 27.9 21.0 81.2 BP Y -1 23.4 0.4 28.7 21.1 81.8 BP Y 0 23.5 0.3 29.3 21.9 82.5 BP % 1-yr 16.9 -43.3 3.8 2.0 146.8 BP % 3-yr 40.4 -76.5 11.8 5.4 328.3 BP % 5-yr 81.5 -91.4 13.8 4.5 1249.5 BP slope 1-yr 1.4 -5.0 0.6 0.4 6.3 BP slope 3-yr 1.3 -5.2 0.6 0.4 4.8 BP slope 5-yr 1.3 -6.8 0.4 0.2 4.8 Innov Y 0 20.1 0.0 59.1 57.8 97.4 Ad Y -9 DOLS (000) 303993.1 0.0 126351.9 58.4 3171682.8 Ad Y -8 DOLS (000) 269733.9 0.0 105140.9 58.4 2930805.0 Ad Y -7 DOLS (000) 255825.4 0.0 98197.9 59.6 2494849.5 Ad Y -6 DOLS (000) 261338.3 0.0 102114.9 20644.8 2931266.2 Ad Y -5 DOLS (000) 252482.4 0.0 101485.2 13215.6 2799303.0 Ad Y -4 DOLS (000) 248359.9 0.0 101501.2 8565.1 2673944.3 Ad Y -3 DOLS (000) 258437.7 0.0 102916.2 9425.0 2924291.9 Ad Y -2 DOLS (000) 243904.1 0.0 98506.4 9428.4 2313153.6 Ad Y -1 DOLS (000) 230025.2 0.0 92045.8 7924.5 1824096.7 Ad 1-yr % 5888.5 -100.0 319.6 7890.2 139735.6 Ad 3-yr % 2543.7 -100.0 266.4 7489.1 47717.0 Ad 5-yr % 1042.2 -100.0 198.7 7094.7 14780.9 Ad 8-yr % 8909.8 -100.0 668.1 -8.0 166374.1 Ad slope 1-yr 60765.6 -784297.4 -6516.2 -13.7 740319.1 Ad slope 3-yr 25022.2 -316068.1 -3187.4 -11.0 224840.4 Ad slope 5-yr 17676.1 -192684.2 -1500.8 -31.3 99642.4 Ad slope 8-yr 15241.3 -132776.7 -966.4 -62.9 85404.5 Y -5 Cash Flow per share: (Earnings 3.42 -3.41 4.60 -28.44 26.06 for Banks and Insurance companies) Y -4 Cash Flow per share: (Earnings 3.70 -3.85 4.91 -9.37 30.85 for Banks and Insurance companies) Y -3 Cash Flow per share: (Earnings 3.94 -2.76 5.30 -18.92 36.27 for Banks and Insurance companies) Y -2 Cash Flow per share: (Earnings 4.42 -2.33 5.70 1.00 40.89 for Banks and Insurance companies) Y -1 Cash Flow per share: (Earnings 4.94 -17.05 5.75 4.09 46.65 for Banks and Insurance companies) Y 0 Cash Flow per share: (Earnings 5.82 -4.30 6.07 4.20 59.65 for Banks and Insurance companies) Y +1 Cash Flow per share: (Earnings 6.49 -4.05 6.80 4.35 62.45 for Banks and Insurance companies) Y +2 Cash Flow per share: (Earnings 6.79 0.50 8.36 4.67 74.80 for Banks and Insurance companies) Y +3 Cash Flow per share: (Earnings 8.05 0.55 8.85 4.80 79.60 for Banks and Insurance companies) CF 1-yr grow % 179.8 -2550.0 6.4 4.9 1836.7 CF 3-yr grow % 108.0 -1533.3 9.8 5.4 1204.3 CF 5-yr grow % 126.7 -648.4 34.1 6.8 1317.3

Table A1 (Continued)

Field Std. Dev Min Mean Median Max CF 1-yr proj % 55.2 -401.7 9.9 7.2 550.0 CF 3-yr grow proj 120.8 -1200.0 45.5 4.3 1300.0 CF 1-yr grow slope 3.4 -7.5 0.4 12.7 56.6 CF 3-yr grow slope 1.7 -3.2 0.3 26.5 28.4 CF 5-yr grow slope 1.5 -2.1 0.4 8.7 28.4 CF 1-yr proj slope 1.5 -16.1 0.7 40.0 12.4 CF 3-yr proj slope 1.0 -4.5 1.0 0.2 9.6 Y -5 Earnings per share 2.46 -5.55 2.77 0.20 19.47 Y -4 Earnings per share 2.66 -6.41 2.99 0.24 23.48 Y -3 Earnings per share 2.89 -6.60 3.28 0.41 27.79 Y -2 Earnings per share 3.22 -7.81 3.57 0.72 31.57 Y -1 Earnings per share 3.94 -21.00 3.55 2.49 36.03 Y 0 Earnings per share 3.85 -7.60 3.64 2.61 40.70 Y +1 Earnings per share 4.15 -7.10 4.23 2.73 45.45 Y +2 Earnings per share 5.14 -1.75 5.75 2.97 58.10 Y +3 Earnings per share 5.36 -1.50 5.89 3.09 61.10 EPS 1-yr grow % 199.9 -700.0 -0.8 3.0 3900.0 EPS 3-yr grow % 116.2 -600.0 9.2 3.5 1400.0 EPS 5-yr grow % 397.3 -1433.3 48.8 4.6 7650.0 EPS 1-yr proj % 177.9 -1650.0 10.1 4.9 2200.0 EPS 3-yr grow proj 350.5 -3400.0 56.7 2.6 2900.0 EPS 1-yr grow slope 1.6 -7.6 0.1 11.4 20.5 EPS 3-yr grow slope 0.8 -3.9 0.1 27.0 4.3 EPS 5-yr grow slope 0.6 -2.4 0.2 9.3 4.2 EPS 1-yr proj slope 1.0 -8.9 0.5 46.7 10.0 EPS 3-yr proj slope 0.8 -0.8 0.8 0.1 7.4 Y -5 Sales/Revenues per share: (Earnings for banks and insurance 59.55 0.43 50.51 0.13 522.27 companies) Y -4 Sales/Revenues per share: (Earnings for banks and insurance 61.60 -0.39 52.27 0.16 540.38 companies) Y -3 Sales/Revenues per share: (Earnings for banks and insurance 64.33 0.69 53.67 0.35 596.80 companies) Y -2 Sales/Revenues per share: (Earnings for banks and insurance 72.40 -0.78 56.78 0.60 773.25 companies) Y -1 Sales/Revenues per share: (Earnings for banks and insurance 77.69 0.58 57.51 30.92 867.70 companies) Y 0 Sales/Revenues per share: (Earnings for banks and insurance 83.25 0.40 58.32 32.73 909.10 companies) Y +1 Sales/Revenues per share: (Earnings for banks and insurance 91.55 0.60 63.67 33.18 977.25 companies) Y +2 Sales/Revenues per share: (Earnings for banks and insurance 102.94 1.05 74.36 35.22 1190.50 companies) Y +3 Sales/Revenues per share: (Earnings for banks and insurance 111.03 1.05 76.32 34.10 1190.50 companies) SPS 1-yr grow % 33.8 -100.0 -0.3 33.4 353.3 SPS 3-yr grow % 38.9 -100.0 5.3 36.5 345.6 SPS 5-yr grow % 59.4 -100.0 20.6 45.1 441.0 SPS 1-yr proj % 42.3 -100.0 4.3 44.0 719.2 SPS 3-yr grow proj 51.9 -100.0 28.1 1.8 766.7 SPS 1-yr grow slope 13.9 -146.8 1.5 6.2 84.7 SPS 3-yr grow slope 10.7 -50.3 1.7 13.4 103.1 SPS 5-yr grow slope 9.0 -39.6 1.8 -- 88.4 SPS 1-yr proj slope 13.6 -17.9 4.4 5.3 203.9 SPS 3-yr proj slope 11.2 -38.4 6.2 26.3 105.7

Table A1 (Continued)

Field Std. Dev Min Mean Median Max Y -5 Dividend per share 0.74 0.02 0.97 0.40 5.50 Y -4 Dividend per share 0.79 0.04 1.05 0.41 6.00 Y -3 Dividend per share 0.85 0.04 1.14 0.52 6.72 Y -2 Dividend per share 0.93 0.04 1.25 1.75 7.72 Y -1 Dividend per share 1.02 0.01 1.36 3.19 8.72 Y 0 Dividend per share 1.21 0.04 1.49 0.80 11.00 Y +1 Dividend per share 1.19 0.04 1.56 0.87 10.30 Y +2 Dividend per share 1.33 0.06 1.78 0.96 12.72 Y +3 Dividend per share 5.30 0.05 2.29 1.09 85.00 DPS 1-yr grow % 50.0 -100.0 6.5 1.2 685.7 DPS 3-yr grow % 111.0 -100.0 46.2 1.3 950.0 DPS 5-yr grow % 213.1 -100.0 104.5 1.4 1800.0 DPS 1-yr proj % 21.1 -100.0 3.2 1.5 60.0 DPS 3-yr grow proj 1081.6 -100.0 112.9 1.7 14566.7 DPS 1-yr grow slope 0.6 -1.9 0.1 6.8 9.6 DPS 3-yr grow slope 0.2 -0.5 0.1 30.6 3.0 DPS 5-yr grow slope 0.2 -0.4 0.1 60.0 1.5 DPS 1-yr proj slope 0.5 -9.3 0.1 5.8 1.1 DPS 3-yr proj slope 1.5 -2.6 0.3 28.8 25.3 Y -5 Book Value per share 17.91 -36.09 19.14 0.08 140.48 Y -4 Book Value per share 19.16 -42.06 20.10 0.10 148.37 Y -3 Book Value per share 20.32 -49.20 22.11 0.10 156.82 Y -2 Book Value per share 22.13 -50.21 22.39 0.08 163.95 Y -1 Book Value per share 24.48 -55.49 23.37 0.12 182.40 Y 0 Book Value per share 27.95 -81.39 24.64 14.63 208.75 Y +1 Book Value per share 28.03 -63.65 25.96 15.82 246.80 Y +2 Book Value per share 32.43 -30.00 32.48 17.05 341.05 Y +3 Book Value per share 35.19 -40.00 33.74 16.50 384.35 BVPS 1-yr grow % 1057.0 -371.6 57.0 17.1 21858.3 BVPS 3-yr grow % 452.6 -324.9 32.3 18.2 9178.2 BVPS 5-yr grow % 329.9 -1190.9 44.7 19.6 4685.7 BVPS 1-yr proj % 129.4 -2500.0 -4.8 23.7 460.0 BVPS 3-yr grow proj 554.3 -10600.0 17.4 25.6 3560.0 BVPS 1-yr grow slope 10.7 -25.9 1.3 3.4 139.0 BVPS 3-yr grow slope 4.8 -14.2 0.9 6.4 50.2 BVPS 5-yr grow slope 3.3 -13.8 1.1 22.5 38.6 BVPS 1-yr proj slope 7.8 -135.2 1.6 7.1 38.1 BVPS 3-yr proj slope 4.9 -36.4 3.5 32.9 62.1 Y -5 Shares Outstanding (millions) 1188.2 22.7 732.9 0.6 10573.0 Y -4 Shares Outstanding (millions) 1169.2 22.7 723.5 0.4 10406.0 Y -3 Shares Outstanding (millions) 1145.8 22.7 717.5 0.7 10061.0 Y -2 Shares Outstanding (millions) 1126.2 10.4 706.8 1.4 10057.0 Y -1 Shares Outstanding (millions) 1112.5 22.0 699.1 2.5 9397.3 Y 0 Shares Outstanding (millions) 4356.6 21.0 1009.7 335.6 68140.0 Y +1 Shares Outstanding (millions) 4445.9 21.0 1019.1 331.3 68950.0 Y +2 Shares Outstanding (millions) 1097.5 19.0 680.6 334.9 10000.0 Y +3 Shares Outstanding (millions) 4961.4 18.0 1053.8 333.0 73000.0 SO 1-yr grow % 161.2 -100.0 8.8 328.0 2847.2 SO 3-yr grow % 158.7 -100.0 6.4 318.8 2847.7 SO 5-yr grow % 160.5 -100.0 6.5 316.1 2847.6 SO 1-yr proj % 19.5 -100.0 -5.2 310.0 15.6 SO 3-yr grow proj 19.7 -100.0 -7.3 314.0 20.8 SO 1-yr grow slope 4027.5 -497.3 313.3 -1.8 65828.0 SO 3-yr grow slope 1187.8 -414.3 85.2 -4.4 19748.6 SO 5-yr grow slope 566.0 -325.6 37.4 -6.2 9404.1 SO 1-yr proj slope 75.4 -947.0 -8.8 -0.8 810.0 SO 3-yr proj slope 268.0 -5204.5 -25.4 -2.8 122.5 Y -5 Sales / Revenues ($mill): (zero 50217.0 0.0 24087.7 -3.0 470171.0 for banks and insurance companies) Y -4 Sales / Revenues ($mill): (zero 49812.0 0.0 24325.3 -3.1 469162.0 for banks and insurance companies) Y -3 Sales / Revenues ($mill): (zero 49656.8 0.0 24558.3 -2.5 476294.0 for banks and insurance companies) 67

Table A1 (Continued)

Field Std. Dev Min Mean Median Max Y -2 Sales / Revenues ($mill): (zero 49018.0 0.0 25001.8 -1.5 485651.0 for banks and insurance companies) Y -1 Sales / Revenues ($mill): (zero 46508.7 0.0 24389.8 -1.6 484000.0 for banks and insurance companies) Y 0 Sales / Revenues ($mill): (zero 42860.4 0.0 23395.1 9134.5 487000.0 for banks and insurance companies) Y +1 Sales / Revenues ($mill): (zero 46380.3 0.0 25242.5 9647.0 500000.0 for banks and insurance companies) Y +2 Sales / Revenues ($mill): (zero 59963.2 0.0 30837.7 10183.5 565500.0 for banks and insurance companies) Y +3 Sales / Revenues ($mill): (zero 54368.6 0.0 28564.6 10846.0 565500.0 for banks and insurance companies) Sls 1-yr grow % 25.6 -100.0 -2.6 10600.0 196.5 Sls 3-yr grow % 35.8 -100.0 1.1 10400.0 254.8 Sls 5-yr grow % 51.4 -100.0 10.6 11122.5 301.7 Sls 1-yr proj % 55.6 -100.0 3.0 13000.0 987.3 Sls 3-yr grow proj 74.3 -100.0 22.1 13400.0 1287.0 Sls 1-yr grow slope 4922.8 -45000.0 175.2 0.2 28250.0 Sls 3-yr grow slope 5845.2 -69850.4 -79.4 1.2 21051.9 Sls 5-yr grow slope 4476.2 -49607.3 98.6 5.6 22350.3 Sls 1-yr proj slope 4878.7 -21000.0 1285.7 3.9 60000.0 Sls 3-yr proj slope 7633.1 -49800.0 1825.8 19.9 81600.0 Y -5 Stock Price: from Yahoo! 48771.06 0.02 2392.95 0.00 1009018.83 Finance Y -4 Stock Price: from Yahoo! 70386.49 0.02 3435.89 0.10 1459606.61 Finance Y -3 Stock Price: from Yahoo! 63154.61 0.02 3056.67 53.59 1329306.38 Finance Y -2 Stock Price: from Yahoo! 61860.69 0.08 2994.84 327.50 1305014.42 Finance Y -1 Stock Price: from Yahoo! 59593.08 0.90 2880.34 768.00 1260000.00 Finance Y 0 Stock Price: from Yahoo! 79.40 3.23 72.16 28.36 789.79 Finance Stk % 1-yr 48.9 -100.0 2.4 34.2 319.2 Stk % 3-yr 59.1 -100.0 10.0 43.9 313.0 Stk % 5-yr 123.4 -100.0 94.3 49.5 915.5 Stk slope 1-yr 14.9 -102.5 7.9 45.6 104.8 Stk slope 3-yr 1643.0 -34653.2 -74.1 56.0 141.9 Stk slope 5-yr 1646.4 -119.3 84.9 10.9 34737.0 Timeliness 9.5 1.0 3.6 17.7 188.5 Safety 10.8 1.0 3.1 89.4 224.7 Technical 1.6 1.0 3.1 6.6 29.2 Financial Strength 2.2 1.0 3.8 2.4 28.3 Price Stability 27.0 5.0 63.0 4.8 100.0

*Key: for all variable abbreviations in this dissertation Fam – Familiarity Rep – Overall Reputation Mgt – Perception of Management Inv – Investment Potential Fav – Favorability BP – BrandPower Innov – Culture of Innovation CF – Cash Flow EPS – Earnings per Share SPS – Sales / Revenue per Share DPS – Dividend per Share 68

BVPS – Book Value per Share SO – Shares Outstanding Sls – Sales / Revenue Stk – Stock Price per Share

APPENDIX B:

CORRELATION ANALYSIS

Table A2. 1-Year Sales / Revenue Per Share, % Growth Correlations.

1yr Growth Variable Correlation Coefficient Strength Statistics Count 433 Mean -0.256 Min -100.000 Max 353.263 Range 453.263 Variance 1148.609 Standard Deviation 33.891 Standard Error of Mean 1.629 Pearson Correlations Fam 4Q 11 0.058 Weak Fam 4Q 12 0.057 Weak Fam 4Q 13 0.057 Weak Fam 4Q 14 0.057 Weak Fam 4Q 15 0.056 Weak Fam 4Q 16 0.056 Weak Fam % 1-yr -0.066 Weak Fam % 3-yr -0.078 Weak Fam % 5-yr -0.036 Weak Fam slope 1-yr 0.005 Weak Fam slope 3-yr 0.009 Weak Fam slope 5-yr 0.010 Weak Rep 4Q 11 0.011 Weak Rep 4Q 12 0.013 Weak Rep 4Q 13 0.011 Weak Rep 4Q 14 -0.008 Weak Rep 4Q 15 -0.011 Weak Rep 4Q 16 -0.025 Weak Rep % 1-yr -0.062 Weak Rep % 3-yr -0.074 Weak Rep % 5-yr -0.051 Weak Rep slope 1-yr -0.062 Weak Rep slope 3-yr -0.070 Weak Rep slope 5-yr -0.053 Weak Mgt 4Q 11 0.020 Weak Mgt 4Q 12 0.028 Weak Mgt 4Q 13 0.018 Weak Mgt 4Q 14 0.013 Weak Mgt 4Q 15 -0.012 Weak Mgt 4Q 16 -0.026 Weak Mgt % 1-yr -0.057 Weak Mgt % 3-yr -0.089 Medium Mgt % 5-yr -0.062 Weak Mgt slope 1-yr -0.061 Weak Mgt slope 3-yr -0.097 Strong Mgt slope 5-yr -0.070 Weak Inv 4Q 11 0.019 Weak Inv 4Q 12 0.019 Weak Inv 4Q 13 0.017 Weak Inv 4Q 14 0.005 Weak

Table A2 (Continued)

1yr Growth Variable Correlation Coefficient Strength Inv 4Q 15 -0.017 Weak Inv 4Q 16 -0.027 Weak Inv % 1-yr -0.048 Weak Inv % 3-yr -0.094 Medium Inv % 5-yr -0.056 Weak Inv slope 1-yr -0.048 Weak Inv slope 3-yr -0.099 Strong Inv slope 5-yr -0.068 Weak Fav 4Q 11 0.017 Weak Fav 4Q 12 0.021 Weak Fav 4Q 13 0.016 Weak Fav 4Q 14 0.003 Weak Fav 4Q 15 -0.014 Weak Fav 4Q 16 -0.027 Weak Fav % 1-yr -0.061 Weak Fav % 3-yr -0.092 Medium Fav % 5-yr -0.062 Weak Fav slope 1-yr -0.062 Weak Fav slope 3-yr -0.095 Strong Fav slope 5-yr -0.067 Weak BP 4Q 2011 0.050 Weak BP 4Q 2012 0.047 Weak BP 4Q 2013 0.045 Weak BP 4Q 2014 0.043 Weak BP 4Q 2015 0.042 Weak BP 4Q 2016 0.041 Weak BP % 1-yr -0.077 Weak BP % 3-yr -0.103 Strong BP % 5-yr -0.052 Weak BP slope 1-yr -0.013 Weak BP slope 3-yr -0.016 Weak BP slope 5-yr -0.020 Weak Innov 1Q 16 -0.041 Weak Innov 2Q 16 -0.025 Weak Innov 3Q 16 -0.017 Weak Innov 4Q 16 -0.027 Weak

Table A3. 3-Year Sales / Revenue Per Share, % Growth Correlations.

SPS 3yr Growth Variable Correlation Coefficient Strength Statistics Count 430 Mean 5.923 Min -100.000 Max 345.571 Range 445.571 Variance 1669.280 Standard Deviation 40.857 Standard Error of Mean 1.970 Pearson Correlations Fam 4Q 11 0.080 Medium Fam 4Q 12 0.083 Medium Fam 4Q 13 0.085 Medium Fam 4Q 14 0.086 Medium Fam 4Q 15 0.083 Medium Fam 4Q 16 0.082 Medium Fam % 1-yr -0.084 Medium Fam % 3-yr -0.103 Strong Fam % 5-yr -0.034 Weak Fam slope 1-yr -0.013 Weak Fam slope 3-yr -0.007 Weak Fam slope 5-yr 0.027 Weak Rep 4Q 11 0.010 Weak

Table A3 (Continued)

SPS 3yr Growth Variable Correlation Coefficient Strength Rep 4Q 12 0.020 Weak Rep 4Q 13 0.017 Weak Rep 4Q 14 0.003 Weak Rep 4Q 15 -0.002 Weak Rep 4Q 16 -0.006 Weak Rep % 1-yr -0.032 Weak Rep % 3-yr -0.064 Weak Rep % 5-yr -0.032 Weak Rep slope 1-yr -0.019 Weak Rep slope 3-yr -0.046 Weak Rep slope 5-yr -0.029 Weak Mgt 4Q 11 0.011 Weak Mgt 4Q 12 0.025 Weak Mgt 4Q 13 0.014 Weak Mgt 4Q 14 0.007 Weak Mgt 4Q 15 -0.009 Weak Mgt 4Q 16 -0.029 Weak Mgt % 1-yr -0.084 Medium Mgt % 3-yr -0.105 Strong Mgt % 5-yr -0.071 Weak Mgt slope 1-yr -0.087 Medium Mgt slope 3-yr -0.092 Medium Mgt slope 5-yr -0.062 Weak Inv 4Q 11 -0.020 Weak Inv 4Q 12 -0.010 Weak Inv 4Q 13 -0.015 Weak Inv 4Q 14 -0.023 Weak Inv 4Q 15 -0.037 Weak Inv 4Q 16 -0.052 Weak Inv % 1-yr -0.075 Weak Inv % 3-yr -0.097 Strong Inv % 5-yr -0.048 Weak Inv slope 1-yr -0.070 Weak Inv slope 3-yr -0.079 Weak Inv slope 5-yr -0.056 Weak Fav 4Q 11 -0.001 Weak Fav 4Q 12 0.011 Weak Fav 4Q 13 0.005 Weak Fav 4Q 14 -0.006 Weak Fav 4Q 15 -0.018 Weak Fav 4Q 16 -0.031 Weak Fav % 1-yr -0.071 Weak Fav % 3-yr -0.095 Strong Fav % 5-yr -0.055 Weak Fav slope 1-yr -0.064 Weak Fav slope 3-yr -0.078 Weak Fav slope 5-yr -0.052 Weak BP 4Q 2011 0.064 Weak BP 4Q 2012 0.065 Weak BP 4Q 2013 0.066 Weak BP 4Q 2014 0.065 Weak BP 4Q 2015 0.064 Weak BP 4Q 2016 0.061 Weak BP % 1-yr -0.096 Strong BP % 3-yr -0.127 Strong BP % 5-yr -0.054 Weak BP slope 1-yr -0.028 Weak BP slope 3-yr -0.011 Weak BP slope 5-yr 0.008 Weak Innov 1Q 16 0.011 Weak Innov 2Q 16 0.040 Weak Innov 3Q 16 0.049 Weak

Table A3 (Continued)

SPS 3yr Growth Variable Correlation Coefficient Strength Innov 4Q 16 0.026 Weak

Table A4. 5-Year Sales / Revenue Per Share, % Growth Correlations.

SPS 5yr Growth Variable Correlation Coefficient Strength Statistics Count 424 Mean 25.045 Min -10000.000 Max 14700.000 Range 24700.000 Variance 801570.790 Standard Deviation 895.305 Standard Error of Mean 43.480 Pearson Correlations Fam 4Q 11 0.073 Weak Fam 4Q 12 0.076 Weak Fam 4Q 13 0.080 Medium Fam 4Q 14 0.084 Medium Fam 4Q 15 0.086 Medium Fam 4Q 16 0.088 Medium Fam % 1-yr 0.011 Weak Fam % 3-yr 0.018 Weak Fam % 5-yr 0.011 Weak Fam slope 1-yr 0.050 Weak Fam slope 3-yr 0.075 Weak Fam slope 5-yr 0.081 Medium Rep 4Q 11 0.032 Weak Rep 4Q 12 0.035 Weak Rep 4Q 13 0.036 Weak Rep 4Q 14 0.040 Weak Rep 4Q 15 0.035 Weak Rep 4Q 16 0.038 Weak Rep % 1-yr 0.010 Weak Rep % 3-yr 0.001 Weak Rep % 5-yr 0.012 Weak Rep slope 1-yr 0.018 Weak Rep slope 3-yr 0.002 Weak Rep slope 5-yr 0.015 Weak Mgt 4Q 11 0.034 Weak Mgt 4Q 12 0.028 Weak Mgt 4Q 13 0.023 Weak Mgt 4Q 14 0.041 Weak Mgt 4Q 15 0.030 Weak Mgt 4Q 16 0.049 Weak Mgt % 1-yr 0.056 Weak Mgt % 3-yr 0.031 Weak Mgt % 5-yr 0.022 Weak Mgt slope 1-yr 0.086 Medium Mgt slope 3-yr 0.040 Weak Mgt slope 5-yr 0.028 Weak Inv 4Q 11 0.055 Weak Inv 4Q 12 0.048 Weak Inv 4Q 13 0.048 Weak Inv 4Q 14 0.056 Weak Inv 4Q 15 0.051 Weak Inv 4Q 16 0.073 Weak Inv % 1-yr 0.076 Weak Inv % 3-yr 0.030 Weak Inv % 5-yr 0.015 Weak Inv slope 1-yr 0.106 Strong Inv slope 3-yr 0.046 Weak Inv slope 5-yr 0.036 Weak

Table A4 (Continued)

SPS 5yr Growth Variable Correlation Coefficient Strength Fav 4Q 11 0.043 Weak Fav 4Q 12 0.039 Weak Fav 4Q 13 0.038 Weak Fav 4Q 14 0.048 Weak Fav 4Q 15 0.041 Weak Fav 4Q 16 0.056 Weak Fav % 1-yr 0.053 Weak Fav % 3-yr 0.024 Weak Fav % 5-yr 0.022 Weak Fav slope 1-yr 0.077 Weak Fav slope 3-yr 0.032 Weak Fav slope 5-yr 0.029 Weak BP 4Q 2011 0.070 Weak BP 4Q 2012 0.073 Weak BP 4Q 2013 0.076 Weak BP 4Q 2014 0.081 Medium BP 4Q 2015 0.084 Medium BP 4Q 2016 0.089 Medium BP % 1-yr 0.026 Weak BP % 3-yr 0.016 Weak BP % 5-yr 0.006 Weak BP slope 1-yr 0.104 Strong BP slope 3-yr 0.097 Strong BP slope 5-yr 0.087 Medium Innov 1Q 16 0.039 Weak Innov 2Q 16 0.053 Weak Innov 3Q 16 0.057 Weak Innov 4Q 16 0.050 Weak

APPENDIX C:

SECTOR CHAID DECISION TREES

Figure A1. CHAID Decision Tree: Consumer Staples Sector.

Figure A2. CHAID Decision Tree: Industrials Sector.

Figure A3. CHAID Decision Tree: Technology Sector.

APPENDIX D:

PERMISSION LETTER

September 27, 2018

Mr. Brad Puckey 206 Grace Manor Dr. Coraopolis , PA 15108

Subject: Permission to use CoreBrand Data

Dear Brad,

This letter grants you a limited, non-exclusive, non-transferable right to utilize the research database (“Data”) known as the CoreBrand Index® owned by Brandlogic Corp. D/B/A Tenet Partners (“Tenet Partners”) solely to use the Data in your doctoral dissertation with the University of South Florida (“Dissertation”).

As consideration, you will give Tenet Partners and CoreBrand Data Science attribution and recognition of its ownership of the Data in the Dissertation. The use of the Data is limited to the Dissertation only and no other rights are conveyed. In addition, you agree to not use the Data or any of Tenet Partner’s trademarks (“Trademarks”) or other use beyond the publication of your Dissertation nor for any commercial use, and you shall not grant any rights to use the Data or the Trademarks to any other parties. Tenet Partners reserves all rights in and to the Data and Trademarks not expressly granted to you, as well as the right to terminate the rights granted hereunder in the event you breach or abuse your rights under this agreement or any other agreements between you and Tenet Partners. This agreement does not replace or augment any other agreements currently active between you and Tenet Partners.

We are excited to see the culmination of your doctorial work in what will prove to be a

77 fascinating exploration and demonstration of the CoreBrand Index and its potential link to better understanding intangible assets of corporations.

Sincerely,

Hampton Bridwell CEO and Managing Partner Tenet Partners 122 W. 27th Street, 9th Floor New York, NY 10001

ABOUT THE AUTHOR

Brad A. Puckey has worked for CoreBrand/Tenet Partners for nearly three decades as the head of the company’s analytics practice. He was there at the beginning of the company’s analytics efforts; he built the CoreBrand Index® database and created the CoreBrand Analysis, the company’s flagship analytic tools. The data created was the key to this dissertation. The

CoreBrand Analysis has been employed by many Fortune 500 companies, including General

Electric, Cisco Systems, Dow Chemical, and many, many others. The data has been used to create brand valuation, communication return-on-investment, and other custom models for clients.

Brad graduated from Indiana University of Pennsylvania with a Bachelor of Arts in

Economics, where he was involved in student government and a member of the Pi Kappa Phi fraternity. Prior to attending college, Brad was a member of the United States Army, where he served as an Electronic Warfare Signals Intelligence (SIGINT) Analyst. He served tours at the

National Security Agency at Ft. Meade, MD; A Company, Military Intelligence Battalion (Low

Intensity) Soto Cano Airbase, Honduras; and 1st Infantry Division (the Big Red One) at Ft. Riley,

KS. [email protected] 203-722-5910