www.hbrreprints.org

Separate the facts from the fads: A groundbreaking, five- What Really Works year study reveals the must- have management practices that truly produce superior by Nitin Nohria, William Joyce, and results. Bruce Roberson

Included with this full-text Harvard Business Review article:

1 Article Summary The Idea in Brief—the core idea The Idea in Practice—putting the idea to work

2 What Really Works

13 Further Reading A list of related materials, with annotations to guide further exploration of the article’s ideas and applications

Reprint R0307C

What Really Works

The Idea in Brief The Idea in Practice New management ideas heat up and fizzle PRIMARY PRACTICES SECONDARY PRACTICES out—seemingly overnight. So how can you To excel at the four primary management Excel at any two of these secondary manage- tell which ones are critical for outperform- practices, consider these guidelines: ment practices. ing your competitors? A groundbreaking study of 200 management techniques re- Strategy: Talent: veals surprising results: Most techniques Build your strategy on deep knowledge of Achieve deep bench strength. It’s cheaper have no direct impact on superior business your target customers and company’s capabil- and more reliable to develop stars than to buy performance. What does? Mastery of busi- ities. Clearly and consistently communicate them. Create top-of-the-line training pro- ness basics. that strategy to employees, customers, and grams to retain skilled managers. Give them shareholders. Refine your strategy only in re- challenging, intriguing jobs. To sustain superior performance, you have sponse to marketplace changes—new tech- Leadership: to excel at four primary management prac- nologies or government regulations, for ex- Successful companies’ leaders are committed tices—strategy, execution, culture, and ample. structure—and any two of four secondary to the business. They reach out to front lines, practices—talent, leadership, innovation, Example: forging connections with people at all levels. and mergers and partnerships. Remaining laser-focused on its strategy They seize opportunities before competitors year after year, Dollar General sets rock-bot- do and tackle problems early. Also, such com- The key to this 4 + 2 formula is not which tom prices and selects only merchandise panies’ board members have a financial stake technique you choose within each practice, appealing to its core customers at the low in the firm’s success and a solid understanding but how well and consistently you stick end of the market. of the industry. with it. There’s no recipe to follow. But the most enduringly successful companies in Execution: Innovation: the study—those delivering a 10-fold re- Streamline operational processes essential to Lead your industry with breakthrough innova- turn to investors over a 10-year period— consistently meeting—not exceeding—cus- tions—even if that means cannibalizing exist- clearly demonstrated hallmarks that any tomer expectations. Eliminate waste to in- ing products. Use new technologies to en- company can follow. crease productivity 6% to 7% annually. hance all operations, not just product- development processes. Culture: Hold managers and employees, individuals Mergers and partnerships: and teams to unyielding performance expec- Enter only new businesses that leverage exist- tations. Link pay to specific goals—and raise ing customer relationships and complement the bar every year. Withhold rewards when your core strengths. Forge partnerships that targets are missed. State company values best use both companies’ talents. Develop a clearly and forcefully. systematic way of identifying, screening, and closing such deals. Structure: Create a fast, flexible, and flat structure that re- duces bureaucracy and simplifies work. Shat- ter departmental boundaries that prevent in- formation sharing and cooperation. Look to middle managers’ and employees’ dedication and inventiveness—not executives’ bril- liance—for your company’s future. © 2003 BY HARVARD BUSINESS SCHOOL PUBLISHING CORPORATION. ALL RIGHTS RESERVED. BUSINESS SCHOOL© 2003 BY HARVARD PUBLISHING CORPORATION. page 1

Separate the facts from the fads: A groundbreaking, five-year study reveals the must-have management practices that truly produce superior results.

What Really Works

by Nitin Nohria, William Joyce, and Bruce Roberson

The dot-com boom of the 1990s had changed that outperformed their industry peers ex- the rules of business forever, it seemed; all you celled at what we call the four primary man- needed was a sexy IPO, cold nerve, and the agement practices—strategy, execution, cul- magic carpet of momentum trading. But even ture, and structure. And they supplemented as entrepreneurs and venture capitalists were their great skill in those areas with a mastery dismissing traditional business models as anti- of any two out of four secondary management quated and conventional business wisdom as practices—talent, innovation, leadership, and old school, we found ourselves wondering if mergers and partnerships. they were right. For years we had watched We learned, for example, that it doesn’t re- new management ideas come and go, passion- ally matter if you implement ERP software or a ately embraced one year, abruptly abandoned CRM system; it matters very much, though, the next. “What really works?” we wondered. that whatever technology you choose to imple- Our curiosity prompted us to undertake a ma- ment you execute it flawlessly. Similarly, it jor, multiyear research effort in which we care- matters little whether you centralize or decen- fully examined more than 200 well-estab- tralize your business as long as you pay atten- lished management practices as they were tion to simplifying the way your organization employed over a ten-year period by 160 com- is structured. We call the winning combination panies. the 4+2 formula for business success. A com- Our findings took us quite by surprise. Most pany that consistently follows this formula has of the management tools and techniques we better than a 90% chance of sustaining supe- studied had no direct causal relationship to su- rior business performance. perior business performance. What does mat- The 160 companies in our study—which we ter, it turns out, is having a strong grasp of the call the Evergreen Project—were divided into business basics. Without exception, companies 40 quads, each comprising four companies in a © 2003 BY HARVARD BUSINESS SCHOOL PUBLISHING CORPORATION. ALL RIGHTS RESERVED. BUSINESS SCHOOL© 2003 BY HARVARD PUBLISHING CORPORATION. harvard business review • july 2003 page 2

What Really Works

narrowly defined industry. The companies in The eight essential management practices each quad began the study period (1986 to we cite are not new, nor is their importance 1996) in approximately the same fiscal condi- particularly surprising or counterintuitive. But tion. Yet their fortunes differed dramatically implementing our formula for success is not as over the decade. One company in each four- simple as it sounds. Companies can all too eas- some emerged as a winner—it consistently out- ily forget or ignore the basics, as we saw in the performed its peers in the industry throughout waning years of the last century. And succeed- our study period; one a loser—it consistently ing at the eight business practices can be hard underperformed against its competitors; one a work. Maintaining a laserlike focus on strategy climber—it started off poorly but dramatically alone, year in and year out, can be grueling. improved its performance once it applied the Yet the winning companies in our study were 4+2 formula; and one a tumbler—it began the running full tilt on six tracks at once—impres- decade in good shape then fell far behind. sive when you consider that a single misstep Over the ten-year period, investors in the win- on any of the six can be fatal. Indeed, some of ning companies saw their money multiply the companies that were deemed winners dur- nearly tenfold, with total returns to sharehold- ing our ten-year research period have since ers of 945%. By contrast, the average loser pro- stumbled in one dimension or another—for in- duced only 62% in total returns to sharehold- stance, Dollar General lost its focus on the val- ers over the decade. (For more on our ues in its culture and, as a result, recently had methodology, see the sidebar “The Evergreen to restate its earnings. It’s much easier to be a Project: Our Research.”) tumbler than it is to remain a winner. Our re- Winners, losers, climbers, and tumblers— search found that less than 5% of all publicly with startling consistency, their fortunes traded companies maintain a total return to marched in lockstep with how well they per- shareholders greater than their industry peers formed on the 4+2 practices. Consider how for more than ten years. And so, it seems, there Tennessee-based retailer Dollar General, a win- is value in being reminded from time to time ner in our study, fared during our research pe- what really works. riod compared to Kmart. (The other compa- nies in their quad were Target and the Excel at Four Primary Practices Limited.) Both companies were in roughly the The primary management practices—strat- same financial shape in 1986, but Dollar Gen- egy, execution, culture, and structure—repre- eral grew steadily, showing healthy profits year sent the fundamentals of business. But what after year. Meanwhile, Kmart floundered, its does it mean to excel in these areas? There are market share plummeting from 30% to 17% be- myriad tools and techniques available to help tween 1990 and 2000. (We confirmed our find- executives master these practices. To improve ings in the five years following the study pe- execution, for example, leaders can employ riod.) Both companies’ performance was TQM, Kaizen, or Six Sigma, among others. directly linked to whether or not they adhered The conventional wisdom about what works to the 4+2 formula. In the strategy practice, for best shifts with the times. Our research shows example, Dollar General never wavered from that while such tools and techniques are help- its focus, which was to provide quality prod- ful and even necessary in streamlining execu- ucts at a low price to low- and fixed-income tion, for instance, or developing strategy, there consumers. Kmart, by contrast, couldn’t seem is no single, obvious choice that will bring a Nitin Nohria is the Richard P. Chap- to decide whether it was focusing on low- or company success. There are, however, hall- man Professor of Business Administra- middle-income consumers. What’s more, it got marks of effective strategy, execution, culture, tion at Harvard Business School in Bos- distracted by a major foray into specialty retail- and structure—which virtually all of our 40 ton. William Joyce is a professor of ing, moving even further from its core custom- winners demonstrated for ten solid years. strategy and organizational theory at ers. At the same time, Kmart was trying to That’s no small accomplishment, especially Dartmouth College’s Tuck School of compete with Wal-Mart on price—a losing bat- given the limited resources companies have Business in Hanover, New Hampshire. tle and in direct conflict with the organization’s and the unpredictable pressures they face. Bruce Roberson is the executive vice effort to go upmarket. (For an overview of how president of marketing and sales at much value the companies in our study re- Strategy Safety-Kleen in Texas and was a partner turned to their shareholders over the ten-year Devise and maintain a clearly at McKinsey & Company in Dallas. period, see the exhibit “How They Fared.”) stated, focused strategy. harvard business review • july 2003 page 3

What Really Works

The Evergreen Project: Our Research

The Evergreen Project began in 1996 made sure that as of 1986, the start of mation on the 160 companies was col- and lasted five years. It grew from our our ten-year study period, the four lected and read by coders trained to shared obsession with two questions: companies in each industry group score each organization on all 200-plus Why do some companies consistently were reasonably equivalent—similar practices on a scale of 1 (poor relative to outperform their competitors? And to one another in scale, scope, finan- peers) to 5 (excellent relative to peers). which of the hundreds of well-known cial numbers, TRS, and apparent fu- We verified the reliability of the survey business tools and techniques can help ture prospects. by obtaining additional information a company be great? We decided to Although they began the study pe- from dozens of people familiar with the carry out a search for evergreen busi- riod as peer businesses in their own in- companies—knowledgeable outsiders, ness success. The project involved dustries, the companies soon parted senior executives, and former execu- more than 50 leading academics and ways. We classified the four in each in- tives who had been present during the consultants using well-accepted re- dustry to represent four archetypes: study period. search tools and procedures to iden- winners, climbers, tumblers, and los- 2. We pursued in-depth studies of tify, collate, and analyze the experi- ers. Winners outperformed their peers several of the management practices ences of 160 companies over a ten-year in TRS during both the first and sec- that we had concluded played a major period. ond five-year periods. Climbers lagged role in enhancing or weakening a com- We selected hundreds of businesses behind their peers in the first period pany’s performance. This second set of that varied in terms of their total re- but moved up in the second. Tumblers studies, many of which were done at turn to shareholders (TRS). Respond- outdid their peers during the first pe- our request by academic experts, al- ing to concerns from some managers riod and faltered in the second. Losers lowed us to verify and extend the larger who view TRS as irrational and prefer scored lower than their peers through survey findings. In each case, though, to be measured by their operating re- both five-year periods. the experts had to test their ideas on sults, we conducted a rigorous analysis By simultaneously studying compa- the same 160 companies included in of the financial statements of all the nies whose performance changed, for our study. companies in our study. We found that better or for worse, we were able to 3. We collected and analyzed hun- the winning companies as measured separate cause and effect. We could dreds of documents concerning these by TRS were also winners when com- identify which management practices companies—newspaper and magazine pared against almost every other actually worked. In other words, we articles, business-school case studies, meaningful measure. Since an individ- could conclude that improving on spe- government filings, analysts’ reports. ual company’s TRS may reflect not so cific practices guarantees a company’s Each company accumulated a stack of much its own performance as the state superior performance—and that fum- paper three inches high, adding up to of its industry, our research compared bling at those practices is bound to 60,000 documents filling 50 storage a company’s TRS with that of its peers worsen performance. Our study used boxes. Supervised by William Joyce, 15 within the same industry. three distinct methodologies to deter- graduate students at Brigham Young From the initial list of companies, mine which management practices University’s business school coded the we selected 160 for detailed study. The truly influence a company’s perfor- documents. This third data collection vast majority had market capitaliza- mance: included market-shaping information, tions between $100 million and $6 bil- 1. We began with a survey methodol- such as the opinions of analysts and lion. We left out failing organizations ogy. We identified more than 200 man- journalists. (This sort of buzz or conver- as well as big conglomerates with di- agement practices that were thought to sation has a huge impact on investors’ verse businesses that could not be influence business success—broad perceptions and thus on every public meaningfully compared with one an- areas such as strategy, innovation, and company’s stock price.) The data from other. We divided the 160 into 40 business processes; and specific prac- the coding process further verified the groups, each comprising four compa- tices including 360-degree feedback, results of the first two sets of analyses. nies in one narrowly defined industry. supply chain management, and the use To keep the playing field level, we of intranets. All publicly available infor-

harvard business review • july 2003 page 4

What Really Works

You can succeed by competing on low prices, tional department store experience. Its value top quality, or great service. And it doesn’t proposition, “psychic comforts at value prices,” matter whether your strategic direction comes is manifest to customers in the form of stores from the CEO, a consultant, or a collaborative that are bright and clean, easy to navigate, and executive team. The key to achieving excel- well stocked with quality products and unique, lence in strategy, whatever you do and how- higher-end merchandise from designers like ever you approach it, is to be clear about what Michael Graves and Todd Oldham. Target your strategy is and consistently communicate chose a clear, viable strategy and stuck with it. it to customers, employees, and shareholders. Now compare Target’s consistency with the It begins with a simple, focused value proposi- Limited empire, a retail winner-turned-tum- tion that is rooted in deep, certain knowledge bler that lost its focus on lifestyle-based fashion about your company’s target customers and a concepts. The company’s branded stores origi- realistic appraisal of your own capacities. nally sold very different merchandise, and Dollar General, for instance, consistently shoppers knew what to expect from each. Ex- sold quality products at low prices to the low press was designed for hip singles, the Limited end of the market. It located its stores in small targeted suburban mothers, and Lerner (which towns and low-income urban areas, priced was part of the Limited’s stable of brands until items at rock bottom, and carefully selected its 2002) served budget-minded career women. merchandise with its core customers in mind. But by the early 1990s, the different stores Target, a climber in our study, has risen to were selling many of the same items, putting become the nation’s second-largest discounter them in direct competition with one another behind Wal-Mart. The company’s climb is best and confusing customers. understood in terms of its leaders’ ability to And then there’s Kmart. It struggled misera- clearly define and establish a highly focused bly throughout the years of our study. Succes- strategy: Provide good value within a tradi- sive CEOs tried to devise strategies that would make the company more competitive, but all of them lacked clarity and consistency. Kmart had always targeted low- and middle-income How They Fared consumers, but when Wal-Mart and Dollar General began to eat away at this clientele, Adherence to the 4+2 formula for business success can have a Kmart decided to pursue a more affluent, fash- significant impact on a company’s fortunes. As the chart shows, ion-conscious consumer. That led to deals with the winners in our study generated the highest total returns to Martha Stewart and Kathy Ireland—but it also shareholders throughout the decade represented in our research prompted Kmart’s disastrous detour into spe- (1986 to 1996). If an individual invested $1 in a portfolio of winning cialty retailing. At the same time, Kmart companies, he or she would have received approximately $11 by fudged its focus because it couldn’t resist the the end of the ten years. If that person invested $1 in the losing urge to go head-to-head with Wal-Mart, cutting companies, he or she would have received only $1.50. prices on thousands of items. Wal-Mart, as usual, refused to be undersold, so Kmart’s price 12 cuts failed to deliver new customers and sim- Winners ply reduced the company’s earnings. 10 Staying clear on strategy means companies need to be careful how they pursue growth. Ex- 8 ecutives are often tempted to seize any oppor- tunity to expand, sometimes pushing their Business School Publishing All rights reserved.Corporation. 6 companies into unfamiliar territory as a result. vard But moving into areas unrelated to the core 4 Tumblers Climbers business inevitably creates strategic drift. Con- fusion reigns, performance falters, profits evap- 2 orate. Our evergreen winners set aggressive

Shareholder return in dollars Losers growth goals—indeed, they grew twice as fast 0 pyright © 2003 Har

Co as the average company in their industries. But 1986 1988 1990 1992 1994 1996 their primary aim was to grow the core busi- Source: Compustat, Evergreen team analysis ness while at the same time expanding only harvard business review • july 2003 page 5

What Really Works

into related markets. a revamped distribution system. The people Over time, ancillary businesses can become closest to the customers—the store managers part of the core, allowing companies to gradu- and employees—received inconsistent mes- ally shift focus as market demands change. sages from the top team and poor support in After all, while you need to stay clear on strat- trying to implement operational and techno- egy—and the essence of what you do will logical changes. Vendors and customers contin- change little over time—you still need to be ued to complain about shabby store displays able to fine-tune your focus in response to new and the fact that Kmart rarely discontinued technologies, social trends, or government reg- items that didn’t sell; unpopular merchandise ulations. Wal-Mart, for instance, stayed focused would languish on the shelves while hot items on providing everyday value to consumers and were frequently out of stock. has continued to grow its core business. Mean- By contrast, Dollar General regularly and while, it has also expanded into new and re- ruthlessly reviewed every stockkeeping unit. lated businesses, like Sam’s Clubs, and into On average, it replaced 150 to 200 items yearly. new geographies, like the United Kingdom. The company used sophisticated information technology at all its stores to accelerate the Execution checkout process and to manage inventory Develop and maintain flawless scrupulously. And it continually tweaked its operational execution. operations. For instance, former CEO Cal As with strategy, it’s not what you execute that Turner, Jr., doubled the amount of space in the matters but how. We found no relationship be- company’s distribution centers, thereby reduc- tween the degree to which a company em- ing the number of runs the retailer’s drivers braced outsourcing, for instance, and its finan- would have to make, and called for a redesign cial performance. Nor did success hinge on the of Dollar General’s stores. They now boast bet- extent to which a company invested in specific ter merchandise-display systems, wider aisles, ERP, CRM, or supply chain management tech- and a brighter, cleaner look. nologies and systems. That’s not to say these Winning companies are realistic. They recog- tools and techniques aren’t useful or produc- nize that there is no way they can outperform tive; it’s just that embracing them won’t neces- their competitors in every facet of operations. So sarily catapult your company to the head of they determine which processes are most impor- your industry. Disciplined attention to opera- tant to meeting their customers’ needs and focus tions is what really counts. their energies and resources on making those To be a steady winner, a company must in- processes as efficient as possible. They take the crease its productivity by about twice the in- same critical eye to product and service quality as dustry’s average. During our research period, well. Evergreen winners deliver offerings that the mean productivity growth across all indus- consistently meet customers’ expectations, and tries was about 3% per year; the winners in our they’re very clear about the standards they have study increased their productivity by 6% to 7% to meet. But they don’t necessarily strive for per- every year. New technologies play a role in fection—unless perfection is explicit in their stra- productivity improvements, but such invest- tegic value proposition, as it is at Federal Express ments must always be judged by whether or and Tiffany. In fact, fully one-third of our win- not they significantly lower costs or boost out- ning companies offered only average product put. Indeed, a hot new technology will not au- quality. Which goes to show that many custom- tomatically enhance a business’s performance ers don’t care about a level of quality that goes any more than steroids can instantly turn ordi- beyond their needs and desires; they won’t neces- nary athletes into gold medalists. sarily reward you for exceeding their expecta- Kmart suffered from an inability to execute tions. They will, however, punish you severely if from the very start of the decade covered by you don’t meet their expectations. You tumble our research. Wal-Mart and Target had raised quickly when you fail on execution. the bar on store design, product availability, and customer service, and Kmart CEO Joseph Culture Antonini knew his company needed to catch Develop and maintain a up. And yet the retailer was never able to fulfill performance-oriented culture. Antonini’s vision of clean, attractive stores and In some quarters of the business world, cul- harvard business review • july 2003 page 6

What Really Works

ture is still considered soft—it’s not taken as one from the janitor to executives on the top seriously as, say, operations. In others, culture team) a sense of ownership over the stores. is considered important, but the emphasis is Rather than insist that each outlet stock identi- on making the work environment fun based cal merchandise and conform to a prescribed on the theory that when employees enjoy layout, Home Depot gives those responsibili- themselves they’re more likely to remain loyal ties to store managers. The practice is some- to the company. what inefficient financially, but it makes the as- Our study made it clear that building the sociates’ work more interesting, exciting, and right culture is imperative, but promoting a rewarding. Kmart’s Antonini, in sharp contrast, fun environment isn’t nearly as important as believed strongly in command-and-control promoting one that champions high-level per- leadership: He put all of the merchandising formance and ethical behavior. In winning and design decisions for all 2,200 Kmart stores companies, everyone works at the highest into the hands of headquarters staff, keeping level. These organizations design and support store employees completely out of the loop. a culture that encourages outstanding individ- Evergreen winners establish and abide by ual and team contributions, one that holds em- clear company values, giving employees a rea- ployees—not just managers—responsible for son to embrace the organization. These are success. Winners don’t limit themselves to best- not vague niceties; winning companies write ing their immediate competitors. Once a com- down their values in clear, forceful language pany has overmatched its rivals in, say, the ef- and demonstrate them with concrete actions. fectiveness of its logistics, it looks outside the Home Depot has identified seven core values, industry. Employees may ask, for instance, including providing excellent customer ser- “Why can’t we do it better than FedEx?” If the vice, creating shareholder value, doing the goal is unreachable, it still represents an oppor- right thing, and giving back to the commu- tunity for high-performing employees and nity. The company has given millions of dol- managers: “If we can’t be the best at logistics, lars in grants to hundreds of organizations in why not outsource it to a partner that can?” four areas: affordable housing, at-risk youth, It should be obvious that the best way to the environment, and disaster preparedness hold people to such high standards is to di- and relief. Team Depot, which is made up of rectly reward achievement. But while nearly thousands of associates, reinforces the com- 90% of the winning companies in our study mitment by pulling together volunteers to, tightly linked pay to performance, only 15% of for instance, rehabilitate housing for home- the losers did the same. The winners were scru- less and low-income families, build safe play- pulous in setting specific goals, raising the bar grounds, and run clinics to educate consum- every year, and enforcing those benchmarks. ers in dealing with emergencies. No bonuses, stock options, or other rewards were given when targets were missed. And the Structure pay-for-performance commitment extended to Build and maintain a fast, the very top of the organization. During the flexible, flat organization. period of our study, officers at steelmaker There’s nothing wrong with bureaucracy per se. Nucor—a company that we classified as a win- Procedures and protocols are necessary for any ner—were rewarded largely through perfor- organization to function well. But too much mance-based bonuses. Their base salaries were red tape can impede progress, dampen employ- lower than those in the industry as a whole. ees’ enthusiasm, and leach their energy. Win- They had no employment contracts, retire- ning companies trim every possible vestige of ment programs, or annuities. And the amount unnecessary bureaucracy—extra layers of man- of their bonuses depended on that year’s re- agement, an abundance of rules and regula- turn on stockholders’ equity. tions, outdated formalities. They strive to make To complement any financial rewards, win- their structures and processes as simple as pos- ning companies develop programs that recog- sible, not only for their employees but also for nize people’s achievements and offer them op- their vendors and customers. portunities to use their talents. Home Depot, That said, no particular organizational for example, has gone to great lengths to give structure separated the winners in our study associates (a term universally applied to every- from the others. It made little difference harvard business review • july 2003 page 7

What Really Works

whether the companies were organized by department heads and plant managers are ex- function, geography, or product. And it didn’t pected to be out in the shop on a regular basis, much matter whether or not they gave their not just listening to problems but also keeping business units P&L responsibility or their new an eye out for ideas, technical developments, businesses permission to adopt structures and or new practices that might have wider appli- processes distinct from the corporate norm. cation throughout the company. What did matter was whether the organiza- Winning companies are convinced that tional structure simplified the work. their future rests not on the brilliance of their Dollar General, in its mission to transform a executives but on the dedication and inventive- small family-run enterprise into a modern cor- ness of their middle managers and employees. poration with professional management, never Decision making isn’t bogged down by a developed superfluous layers of bureaucracy— lengthy chain of command, so employees are what Cal Turner used to call “staff infection.” free to create and innovate. But such a struc- Its lean structure enabled it to shift gears ture isn’t easy to maintain; bureaucracy has a quickly—a point of pride in an otherwise con- way of creeping back into any organization. servative corporate culture. Texas-based insurer USAA calls the discipline Nucor confined its management structure of simplifying structure and processes “paint- to four layers—foreman, department head, ing the bridge.” That is, once you’ve finished plant manager, and CEO—as compared to painting a bridge, prudent maintenance re- nine or more layers of management at other quires that you go back to the other side and major steel companies. That streamlined struc- start over. So it is with bureaucracy: Once a ture was possible only because then-CEO Ken company has assessed all its core processes and Iverson and his aides had pushed significant scraped off the bureaucratic barnacles, it’s time power and responsibility down the line to the to begin again. plant managers and on to the foremen and frontline workers. As a result, managers at Embrace Two of Four Secondary Nucor don’t run meetings, write letters, and Practices push paper. They answer questions from front- Many people would argue that among the sec- line teams and provide them with support and ondary practices of evergreen business suc- resources when they are asked—and only cess—talent, innovation, leadership, and when asked, since the teams are assumed to be mergers and partnerships—excellence in at able to resolve most problems on their own. least talent and leadership is every bit as man- Managers at the steelmaker lead by staying out datory as excellence in each of the four pri- of the way. mary practices. But that’s not the case. The Of course, frontline employees and manag- winning companies in our study comple- ers can make good decisions only if they have mented their strengths in the four primary access to relevant, up-to-date information. But practices with superior performance in any sharing doesn’t come easily, particularly in two of the secondary practices. It didn’t matter large businesses where divisions and depart- which two areas they chose; we didn’t detect ments compete for limited resources. Technical any patterns in the combinations. Perhaps discoveries and best practices are held close to even more surprising, it doesn’t seem to make the vest. Just talking about how valuable any difference if a company excels in all four knowledge sharing is won’t be enough to over- secondary practices rather than just two. come people’s instinct to hoard. The winning There is, apparently, no reward for going be- companies in our study spent considerable yond the 4+2 formula. time, money, and energy on programs and technologies designed to force open the Talent boundaries and get divisions and departments Hold on to talented employees and find more. cooperating and exchanging information—and The best sign we could find that a company it paid off. When he was CEO, Nucor’s Iverson had great talent was the ease with which any regularly toured the divisions, acting as a executives who were lost to competitors could human sponge, absorbing news about the be replaced from within. The winners in our value being generated at different units and study hired chief executives from the outside then disseminating it corporatewide. Nucor’s half as often as the losers did. They seemed to harvard business review • july 2003 page 8

What Really Works

understand that it’s much cheaper to develop people from outside a company. a star than it is to go out and buy one. It’s also more reliable; you’re getting a known quan- Innovation tity. What’s more, worker continuity and com- Make industry-transforming innovations. pany loyalty have taken on far greater impor- What passes for technical achievement in tance post–Internet boom. So the winners most companies—marginal improvements to that chose talent as one of their secondary existing products, for example—would never practices demonstrated a distinct preference satisfy organizations that excel at innovation. for developing and promoting their own stars They’re focused on finding altogether new and an ability to retain their top performers. product ideas or technological breakthroughs A commitment to promote from within is that have the potential to transform their in- meaningless unless the company offers train- dustries. At these companies, innovation isn’t ing and development that can prepare employ- just about turning out new products and ser- ees for new jobs in the company and creates vices; they also apply new technologies to conditions that encourage employees to enroll their internal workings, which can yield huge rather than penalize them for taking time savings and can transform an industry. Innova- away from their jobs. Not long ago, the as- tion also includes the ability to foresee and sumption was that upwardly striving employ- prepare for disruptive events. ees were solely responsible for preparing them- But the interesting thing about this practice selves for higher-level positions. No more. At is that despite voluminous research into which pharmaceutical company Schering-Plough, for structures most effectively encourage innova- instance, between 75% and 80% of vacancies tion, we found no correlation between the are filled from within, and more than 2,000 sources the winners in our study used and the employees per year take production courses. general sources of innovative business ideas. Georgia-based Flowers Foods, one of the larg- Neither internal R&D labs nor external labs, est bakery foods companies in the United neither frontline employees nor management, States, offers not only the usual training and neither customers nor suppliers were necessar- education but also two programs that reinforce ily where winning companies found their key its commitment to employees’ development. innovations. Any one of the winners might The first program prepares employees to be- have relied successfully on one or more of come baking technicians. By providing work- those sources, but none proved essential to the ers with detailed knowledge about operations winners as a group. What the group had in and equipment in its high-tech plants, the common was the ambition to lead the way company prepares them to move off the pro- with major, industry-changing innovations and duction line and into technical roles. In the sec- a willingness to cannibalize offerings, resisting ond program, Flowers sells its delivery routes the temptation to wring every last cent out of to workers who have the requisite training and an existing product before introducing another expertise to take them on. The goal is to give to take its place. employees an opportunity to own their own Schering-Plough, for instance, is a confirmed businesses. cannibal. It actively turns its prescription- A talented employee can be just as valuable only medications into lower-priced, over-the- and hard to replace as a loyal customer. Yet counter ones, automatically displacing the many companies that go to great lengths to re- originals. But sales of OTC drugs typically tain a customer won’t lift a finger to hold on to double or triple quickly. At Home Depot, as a skilled, seasoned manager. About half the well, cannibalization is routine. When a winners in our study excelled in the talent store becomes so popular that employees practice, and these companies dedicated major can no longer maintain a customer-friendly resources—including personal attention from atmosphere, the company opens another top executives—to building and retaining an outlet nearby. effective workforce and management team. It Given the copious literature on corporate is a fallacy that companies must choose be- innovation, it might be expected that most of tween promoting from within and hiring out- our winning companies would have excelled at side talent. Winning companies do both; a tal- innovation. In fact, a bare majority did so— ent-rich environment tends to attract able which underscores how difficult this practice harvard business review • july 2003 page 9

What Really Works

is. Innovation is not to be entered into lightly. Another important quality is the leader’s ability to spot opportunities and problems Leadership early. Some leaders rely on intuition. Others Find leaders who are committed create special groups within the organization to the business and its people. assigned to stay abreast of changes in every- It’s no longer fashionable to accord celebrity thing from politics to demographics. Still oth- status to the chief executive, but there are few ers engage outside consultants or academics to events of greater significance to an organiza- watch for changes in the marketplace. Though tion than its selection of a CEO. In a study con- their methods vary, effective leaders help their ducted by one of us, it was shown that CEOs companies remain winners by seizing opportu- influence 15% of the total variance in a com- nities before their competitors do and tackling pany’s profitability or total return to share- problems before they become troublesome holders. To put that into perspective, the same nightmares. Cisco’s John Chambers is a good study found that the industry in which a com- example. He was quick to realize when the In- pany operates also accounts for a 15% variance ternet bubble burst that Cisco would have to in profitability. So the choice of a new chief ex- write off inventory and otherwise restructure ecutive is just as important as the choice of itself. His willingness to react swiftly allowed whether to stay in the same industry or enter Cisco to bounce back much faster than its ri- a new one. vals did. As vital as a company’s senior leadership No discussion of leadership would be com- team can be, we found that some common be- plete without mentioning the board of direc- liefs about leadership actually had little to do tors, not least because good boards tend to with a company’s becoming and remaining a choose good CEOs. And what defines a good winner. For example, it didn’t matter whether board? Our results suggest that most of the the leader made his or her decisions indepen- current recommendations being championed dently or in collaboration with the top man- by governance-reform advocates don’t matter. agement team. It made little difference Only two characteristics really matter: The whether senior managers relied on quantita- board members should truly understand the tive or qualitative assessments to make key de- business, and they should be passionately com- cisions. Nor was there any correlation between mitted to its success, which is best accom- the personal characteristics of the CEO— plished by giving members a substantial stake whether he or she was viewed as a visionary or in the company’s financial performance. detail-oriented, secure or insecure, patient or impatient, charismatic or quiet—and a com- Mergers and Partnerships pany’s success. Seek growth through mergers Certain CEO skills and qualities do matter, and partnerships. however. One is the ability to build relation- Innovation is one way to drive growth. Pursuit ships with people at all levels of the organiza- of mergers and partnerships is another. While tion and to inspire the rest of the management many of our companies engaged in some team to do the same. CEOs who present them- merger activity, only a small number (22%) selves as fellow employees rather than masters were able to make this a winning practice. Our can foster positive attitudes that translate into research indicates that companies that do rela- improved corporate performance. When David tively small deals (less than 20% of the ac- Johnson was chief executive at Campbell Soup, quirer’s existing size) on a consistent basis a winner in our study, he constantly sought (about two or three every year) are likely to be ways to reach out to employees. He organized more successful than organizations that do rallies where he sometimes donned a red-and- large, occasional deals. The winners in our white apron and chef’s hat. He led managers study appeared to make better choices: In the on wilderness trips to build esprit de corps. deals we analyzed, they created value in most Meanwhile, Kmart’s string of CEOs failed to of the deals they struck, generating returns in break from the company’s top-down, strictly three years that exceeded the premium paid. hierarchical culture. Even the best-intentioned By contrast, the losers destroyed shareholder among them made little effort to reach out to value in most of the deals they did. the front line. Winners and climbers shared no single mo- harvard business review • july 2003 page 10

What Really Works

Making 4 + 2 Work for You

Besides identifying the management practices that can significantly affect a company’s performance, we’ve developed a list of behaviors that support excellence in each practice. The practices and accompanying mandates are outlined below.

Primary management practices

Strategy delight your customers, but make sure formance bar. Whatever your strategy, whether it is low never to disappoint them. • Pay psychological rewards in addition to prices or innovative products, it will work • Deliver products and services that con- financial ones. if it is sharply defined, clearly communi- sistently meet customers’ expectations. • Create a challenging, satisfying work en- cated, and well understood by employees, • Put decision-making authority close to vironment. customers, partners, and investors. the front lines so employees can react • Establish and abide by clear company • Build a strategy around a clear value quickly to changing market conditions. values. proposition for the customer. • Constantly strive to eliminate all forms Structure • Develop strategy from the outside in, of excess and waste; improve productiv- Managers spend hours agonizing over based on what your customers, partners, ity at a rate that is roughly twice the in- how to structure their organizations (by and investors have to say—and how they dustry average. product, geography, customer, and so on). behave—not on gut feel or instinct. Culture Winners show that what really counts is • Continually fine-tune your strategy Corporate culture advocates sometimes whether structure reduces bureaucracy based on changes in the marketplace— argue that if you can make the work fun, and simplifies work. for example, a new technology, a social all else will follow. Our results suggest that • Simplify. Make your organization easy to trend, a government regulation, or a holding high expectations about perfor- work in and work with. competitor’s breakaway product. mance matters a lot more. • Promote cooperation and the exchange • Clearly communicate your strategy • Inspire all managers and employees to of information across the whole com- within the organization and to custom- do their best. pany. ers and other external stakeholders. • Empower employees and managers to • Put your best people closest to the ac- • Keep focused. Grow your core business, make independent decisions and to find tion. and beware the unfamiliar. ways to improve operations—including • Establish systems for the seamless shar- Execution their own. ing of knowledge. Develop and maintain flawless opera- • Reward achievement with pay based on tional execution. You might not always performance, but keep raising the per-

Secondary management practices

Talent • Relentlessly pursue disruptive technolo- to spot opportunities and problems Winners hold on to talented employees gies to develop innovative new products early. and develop more. and services. • Appoint a board of directors whose • Fill mid- and high-level jobs with outstand- • Don’t hesitate to cannibalize existing members have a substantial stake in the ing internal talent whenever possible. products. company’s success. • Create and maintain top-of-the-line • Apply new technologies to enhance all op- Mergers and Partnerships training and development programs. erating processes, not just those dedicated Internally generated growth is essential, • Design jobs that will intrigue and chal- to designing new products and services. but companies that can master mergers lenge your best performers. Leadership and acquisitions can also be winners. • Keep senior management actively in- Choosing great chief executives can raise • Enter new businesses that leverage exist- volved in the selection and development performance significantly. ing customer relationships and comple- of people. • Closely link the leadership team’s pay to ment core strengths. Innovation its performance. • When partnering, move into new busi- An agile company turns out innovative • Encourage management to strengthen nesses that make the best use of both products and services and anticipates dis- its connections with people at all levels partners’ talents. ruptive events in an industry rather than of the company. • Develop a system for identifying, screen- reacting when it may already be too late. • Inspire management to hone its capacity ing, and closing deals.

harvard business review • july 2003 page 11

What Really Works

tivation in their determination to buy or join making—for instance, establishing dedicated with other organizations. Some were seeking teams comprised of individuals with the requi- cross-selling opportunities, others wanted site investigative, financial, business, and nego- economies of scale, while still others were sim- tiation skills. Winning companies often have ply chasing market share. What they didn’t do codified principles—lessons drawn from expe- was enter deals in order to diversify into areas rience—that enable them to more consis- far removed from their core business—gener- tently choose the right partners and integrate ally a losing proposition. them quickly. A merger or acquisition makes sense only • • • when the move leverages the buyer’s or seller’s Our research makes it clear why so few compa- existing customer relationships or comple- nies maintain a steady lead. Business success ments both companies’ existing strengths. In requires unyielding vigilance in six manage- 1994, Cardinal Health, an Ohio drug whole- ment practices at once and constant renewal saler, took over Whitmire Distribution, based to stay on top. Falling down is easy; climbing in California. It was Cardinal’s 11th acquisition back up is not. in a decade, and it effectively doubled the com- Nike, for example, was a high flier at the be- pany’s sales. Cardinal had become an industry ginning of our research period but lost sight of leader in quality service, and Whitmire had a the business basics and became a tumbler. In high-quality customer base. The deal allowed its strategy practice, for instance, Nike failed to Cardinal to bring its services to a new set of notice and respond appropriately when the customers, lifting the company into the upper tastes of its target customers—urban teenag- ranks of its industry. ers—shifted from sneakers to casual wear. In As an alternative to an outright acquisition, an attempt to regain market share, the com- some companies enter into partnerships, pany pushed into brand extensions, losing which can yield growth by allowing two com- focus completely. And in its utter dedication to panies to move into new businesses using the unlimited expansion, Nike lost sight of the pri- talents of both, uniquely combined. (Think of mary practice of execution, neglecting to ride Dow Chemical’s partnerships with Asahi Glass herd on workplace efficiency and cost controls. and Owens-Illinois.) Partnerships provide some But cautionary tales aside, we believe our of the same advantages that mergers do and study offers hope. In the hurly-burly of busi- lack many of the disadvantages. Partners aren’t ness competition, managers yearn for clarity, expected to accommodate all of each other’s certainty, and solid directions for success. The idiosyncrasies, for example. They remain sepa- 4+2 formula is intended to provide just that; it rate entities, united in the expectation that tells managers which management practices their individual talents can be combined in a they need to focus on and which they can ig- new business venture that will benefit both be- nore. The formula is a true-north compass that yond what either might have gained alone. works in any business climate. The winners and climbers in our study didn’t treat acquisitions and partnerships casu- Reprint R0307C ally or as one-off deals. They invested substan- To order, see the next page tial financial and human resources in develop- or call 800-988-0886 or 617-783-7500 ing an efficient, ongoing process for deal or go to www.hbrreprints.org

harvard business review • july 2003 page 12

What Really Works

Further Reading ARTICLES Strategic Business Modeling nience, fewer mistakes, time and money sav- Harvard Business Review Article collection ings—and ultra-loyal customers. May 2002 Also consider noncompetitive suppliers who Product no. 1032 use similar resources to serve the same cus- This collection focuses on strategy—a pri- tomers. General Mills yogurt and Land O’Lakes mary practice described in “What Really butter now ride in the same trucks to the Works.” The authors distinguish strategy from same supermarkets, lowering distribution business models. Your business model de- costs for both companies. scribes who your customers are, how you de- Uncovering Hidden Value in a Midsize liver value to them, and how you make Manufacturing Company money. Your strategy explains how you’ll differ by James E. Ashton, Frank X. Cook, Jr., and from rivals—by performing different activities, Paul Schmitz or similar activities differently. You align your Harvard Business Review business model and strategy, then communi- June 2003 cate the resulting message to employees Product no. R0306H through a strategic principle—a pithy phrase capturing your firm’s uniqueness; for exam- These authors affirm the importance of get- ple, Southwest Airlines’ “Meet customers’ ting back to basics—and resisting the siren- short-haul air travel needs at fares competitive song of new management fads. They focus on with the cost of car travel.” midsize companies seeking to grow by 15% to 20% per year. Rather than leaping into new The collection includes “Why Business Models businesses, they recommend making more of Matter” by Joan Magretta, “What Is Strategy?” the businesses you’re currently in and build- by Michael E. Porter, and “Transforming ing a foundation of operational excellence— Corner-Office Strategy into Frontline Action” high performance in all areas contributing to by Orit Gadiesh and James L. Gilbert. customer satisfaction. The Superefficient Company This foundation enables riskier moves later, as by Michael Hammer your company progresses along the strategic Harvard Business Review pathway. This sequence of priorities starts September 2001 with protecting your existing business—and Product no. R0108E proceeds to further penetrating existing mar- Hammer explains how to excel at another pri- kets, entering new markets, and, lastly, diversi- mary practice—execution—by eliminating fying with new products. The article explains waste and duplication to achieve supereffi- how to map your market segments against ciency. Superefficient companies go beyond competitors’ to unearth larger-than-expected To Order streamlining internal processes and cross-unit markets and unanticipated competitors en- collaboration to streamlining processes croaching on “your” markets. This process For Harvard Business Review reprints and shared with other companies. helps you identify which aspects of your busi- subscriptions, call 800-988-0886 or ness are most worth protecting—and how to Many start with their supply chains. For exam- 617-783-7500. Go to www.hbrreprints.org safeguard them. ple, IBM integrated its fulfillment process with For customized and quantity orders of customers’ procurement processes, enabling Harvard Business Review article reprints, customers to enter their own orders and call 617-783-7626, or e-mail check order status. Results? Greater conve- [email protected] page 13