Leadership & Management

The Knowing-Doing Gap

Jeffrey Pfeffer and Robert Sutton explore how companies can turn talk into action, and why promoting internal competition may not help.

November 01, 1999


Most business executives in the United States believe strongly in the virtues of competition, not only between organizations but within them as well. Competition fits the cultural emphasis on individualism in the United States, where a social Darwinist philosophy emphasizes the many benefits of the survival of the fittest. Beliefs about competition are so ingrained that they serve as almost mindless, automatic, but powerful principles for organizing and managing individual behavior. As a consequence, companies do all kinds of things to encourage competition, creating a “race track” where only one person, group, division, or subunit can win.

Management practices that produce internal competition are so common they seem unexceptional. Examples include forced distributions of performance evaluations, so that only some fraction of people can earn the highest evaluation (something that is done in schools also); recognition awards given to individuals, such as employee-of-the-month programs; forced distributions of individual merit raise budgets, so what one person receives another cannot; contests between departments or divisions or even between individuals within units for various prizes; and published rankings of individual or unit performance. Each of these practices creates a zero-sum situation where the success or rewards of one person or department must come at the expense of another. There can be only one Number One, so for someone to be first, others must be ranked lower, including last.

There is no doubt that zero-sum games can inspire people to work hard, and that the individual winners of these internal competitions benefit from their victories. In our research, however, we uncovered case after case where the costs of such individual victories were borne by those people, groups, and units that lost the contests. And these internal competitions didn’t just harm the losers. They harmed everyone who had a stake in the organization.

Myths About Competition We’ve all heard the widely used rationale that “competition brings out the best in us.” Yet, once you start thinking about this assumption, it is obvious that dominating others in a zero-sum contest is just one form of success. Other sorts of success can be achieved without dividing people into winners and losers. Alfie Kohn, in his book No Contest: The Case Against Competition, explained that people can achieve high goals and performance whether or not they are competing against others. “Success and competition are not at all the same thing,” he wrote, noting that he could succeed in writing a book without trying to make it better than anyone else’s book. “Competition need never enter the picture in order for skills to be mastered and displayed, goals set and met.”

Kohn’s exhaustive review of the research literature on the effects of competition in organizations and in schools led him to conclude that “superior performance not only does not require competition; it usually seems to require its absence.” The failure of competition to invariably foster superior performance is readily explained: Trying to do well and trying to beat other people are two different things. What does affect performance, if it isn’t competition? There is a large body of research showing the power of the self-fulfilling prophecy, also called the Pygmalion effect on performance. Independent of skill, intelligence, or even past performance, when teachers believe that their students will perform well, they do. Independent of other factors, when leaders believe their subordinates will perform well, these positive expectations lead to better performance.

Most of these studies show that the self-fulfilling prophecy occurs when teachers or leaders believe, or act as if they believe, that a randomly selected subset of students or employees will perform better in contrast to some average or unknown group. But research done in the Israeli Defense Forces shows that Pygmalion effects can be created without inducing contrasts between high and low performers. When platoon leaders at training camps were convinced that all of the soldiers in their classes had unusually high command potential, there was still a strong Pygmalion effect. This suggests that overall performance of a group can be increased when leaders expect everyone to do well. There is apparently no need to sort people into subgroups of high-status “winners” and low-status “losers” in order to use the power of the self-fulfilling prophecy as a way to enhance performance.

That’s a good thing, because self-fulfilling prophecies are just as powerful in their effects in the opposite direction. When a leader believes that a person lacks skill or motivation, these negative expectations decrease performance. The effects of negative self-fulfilling prophecies help explain why, over the long term, internal competition can hamper performance. Once a person, group, or division is labeled a loser, research suggests that subsequent performance will be worse because leaders and others will unwittingly act to fulfill the poor performance expectations. And the loss of self-worth and motivation felt by those who are treated as losers leads to further decreases in their performance.

This is why quality guru W. Edwards Deming was so vehemently opposed to relative performance evaluations. Deming emphasized that forced rankings and other merit ratings that breed internal competition are bad management because they undermine motivation and breed contempt for management among people who, at least at first, were doing good work. He argued that these systems require leaders to label many people as poor performers even though their work is well within the range of high quality. Deming maintained that when people get unfair negative evaluations, it can leave them “bitter, crushed, bruised, battered, desolate, despondent, dejected, feeling inferior, some even depressed, unfit for work for weeks after receipt of the rating, unable to comprehend why they are inferior.” These insidious effects of negative evaluations are the reason why direct sales organizations such as Mary Kay, Tupperware, and Avon try instead, through rewards and recognition, to praise everyone to success.

Another common rationale for ranking units or people against each other is that such systems are inherently fairer than other incentives for desired behavior. Many organizations have adopted some form of relative performance evaluation in administering both performance management and compensation systems. The theory behind this at first seems correct. Consider: If you, as a salesperson, are on commission or are evaluated based on your absolute sales volume, you can be punished in your performance if the company has product quality problems or is lagging in product design and innovation — even though you have nothing to do with those problems. In contrast, by assessing your performance in comparison to similarly situated salespeople, these extraneous, uncontrollable factors are taken into account. Even if the company has product or manufacturing problems, presumably those problems affect all salespeople equally. So as long as you perform better than your peers, you score well on the relative performance evaluation and will be rewarded and evaluated accordingly.

Yet the strength of the system — its apparent fairness in the face of external factors over which people have no control — is also its weakness. Because people are concerned primarily about their relative ranking, there are incentives for them to avoid helping their peers to improve their performance and, at worst, to undermine or sabotage their peers’ performance. As Deming described it, when these kinds of performance ratings are used, “Everyone propels himself forward, or tries to, for his own good, on his own life preserver. The organization is the loser.”

Many organizations have embarked on substantial knowledge management initiatives, often with huge investments in information technology, in an effort to share performance knowledge. But technology is not the answer, as numerous reports and studies document. Even as companies try to get units to learn from each other, they fail to build “sharing cultures” in which cooperative behavior is rewarded. Thomas Davenport and Laurence Prusak, in their book Working Knowledge, emphasized the importance of rewarding knowledge sharing rather than knowledge hoarding and the importance of building trust — neither of which internal competition typically encourages — to make knowledge management successful. They commented, “Knowledge altruism is real and can be encouraged. It flourishes in organizations that hire nice people and treat them nicely.”

Why Companies Overemphasize Competition The confusion between what it takes to do well in routine physical tasks as opposed to novel intellectual tasks is one reason that people develop misguided beliefs about the positive effects of competition on performance. People in business, particularly men, often draw on analogies from sports to guide their thinking about how work should be organized and rewarded. And for some physical tasks there is evidence that people perform better when racing against an opponent than when racing against a clock. But treating sports as analogous to business is misguided. This is because hundreds of studies show that intellectual tasks that require learning and inventing new ways of doing things are best performed under drastically different conditions than tasks that have been done over and over again in the past.

People are better at learning new things, being creative, and doing intellectual tasks of all kinds when they don’t work under close scrutiny, they don’t feel as if they are constantly being assessed and evaluated, and they aren’t working in the presence of direct competitors. There is a vast amount of evidence that working around others, especially outsiders who are thought to be judging one’s work, enhances performance for tasks that are well learned and that do not require the acquisition of new skills or novel responses. This is called the social facilitation effect. But these same conditions lead to worse performance on tasks that require complex mental processes and attention; the so-called social inhibition effect makes it harder to learn new things or generate new ideas. Related research strongly suggests that competition inhibits learning and creativity because, rather than focusing on the task at hand, people focus their attention too heavily on what competitors are doing and on the reactions of third parties such as leaders and peers. Moreover, when a task is difficult or complex enough that it requires help or sharing ideas with others, internal competition is especially destructive.

Interdependence is another important way that tasks differ. In the racing analogy that is so commonly employed, a person’s speed is almost completely a function of that individual’s own conditioning, ability, stamina, and mental attitude. But interdependence is what organizations are about. Productivity, performance, and innovation result from joint action, not just individual efforts and behavior.

Even Lincoln Electric, widely known for its individual piece-rate systems, depends heavily on incentives for cooperation that have been built into the system from the beginning. These include a profit-sharing system where approximately 50 percent of an employee’s pay depends on the overall success of the enterprise and a performance evaluation system where cooperation is one of the four dimensions that supervisors use to assess employees. So, while cofounder James F. Lincoln glorified in the individual and espoused the virtues of competition, he also was wise enough to understand there was considerable interdependence among employees and that teamwork should be encouraged.

Yet another reason that so many organizations establish dysfunctional internal competition is that leaders and managers, even more so than most people, have achieved their positions by winning a series of competitions, both in their corporations and during their schooling. The traditional leadership model is based on a competitive dynamic that emphasizes winning a contest where one person’s success requires the failure of others. In school, success is defined in terms of class rank — a zero-sum, competitive outcome. Once on the job, the race to triumph over peers is more intense and the odds of winning are even smaller than in school. In consulting firms and investment banks, only a small fraction of each cohort becomes partner. Even once a person has achieved partnership, compensation frequently depends on doing better than other partners. In companies, promotion up the ladder requires winning a tournament against one’s peers.

And so it goes. The people in our society who have the greatest influence over how firms are organized, either because they manage them or they give advice to those who manage them, are also people who are the most heavily trained and rewarded to believe that internal competition is the best way to organize human activity. Yet there is a striking discrepancy between what we know about leadership and what we do in many of the settings in which leadership is presumably taught and learned. Even as we realize that the most important skills for leadership actually entail the ability to work in teams, to collaborate, and to empathize with others, we train and develop leaders in settings that emphasize internal competition.

When Competition Works We don’t want to create the impression that competition is always a bad thing. Organizations such as Southwest Airlines, AES Corporation, and the Men’s Wearhouse are fierce competitors, which is one of the reasons they have been so successful. It is just that their competitive juices are aimed more at external competitors instead of at people from other locations, units, or departments within the firm, or even at peers in the same unit.

Southwest Airlines, for instance, has a collective internal orientation. The compensation system features collective rewards such as profit sharing and stock ownership; individual merit pay is eschewed. The idea of helping one another to turn planes around, to share ideas, and to build a strong, unified corporate culture is emphasized. But as Colleen Barrett, Southwest’s executive vice president for customers, explained: “With all this success and no apparent external enemy, we had begun fighting among ourselves. The appearance of the United Shuttle on the scene rekindled our Ôwarrior culture’ and got people back focused on fighting the competition instead of thinking so much about each other and how they were doing compared to the others.”

The Men’s Wearhouse reduces internal competition by simply not tolerating it when it becomes excessive and destructive. The successful off-price retailer of tailored men’s clothing pays salespeople a base salary along with commissions based on sales. These commissions are obviously individual rewards and could lead to salespeople competing with each other for customers. Part of the firm’s value system and strategy, however, emphasizes team selling, where people help each other in providing customer service. The pressure to work collaboratively starts during employee orientation at the Men’s Wearhouse Suits University. For instance, as part of the training and orientation for wardrobe consultants, Charlie Bresler, the executive vice president for human development, tells them to define their success only “when your teammates, the sales associates, the tailors, and other wardrobe consultants and management people in the store are also successful.”

This talk is backed with action. If someone is writing a lot more sales than other colleagues in a store, say 25 percent more, it is taken as a signal that the person is probably not sharing walk-in traffic but rather is hogging business. Particularly if that person is also not writing high volumes of business per transaction, he or she is reminded about team selling and the expectation that in the company success comes only when everyone in the store is successful. If the behavior persists, the company will fire that individual.

Companies can also reduce internal competitive dynamics by refusing to use measures and compensation systems that emphasize success at the expense of others. Willamette Industries is one of the most successful forest products, building materials, and paper companies as measured by return on equity, return on sales, and shareholder returns. As a fully integrated company, cooperation across units is important for the firm’s success. To encourage such cooperation, Willamette uses no short-term performance bonuses. As Dave Morthland, the vice president for human resources, explained the firm’s compensation philosophy and its relationship to its interdependent structure: “We control the tree literally from the time it’s planted to the time our finished product leaves the shipping dock. That means we’ve got a lot of internal customer relationships. What kind of message does it send in an integrated company, where you want and need good teamwork, if one side of the business is doing really well because their prices happen to be high and you’re paying them big bonuses, and the person on the other side of the aisle is in the down cycle part of the business, but that person may be working harder to keep his costs down and he’s getting no reward for it?” Willamette has few of the transfer price squabbles that bedevil competitors by absorbing management attention and time and slowing down decision making.

Cooperation has somehow developed a bad reputation in many organizations. Because it has triumphed as an economic system, people think that competition within organizations is a similarly superior way of managing. But there are many instances where internal competition has real consequences that hurt real people and real organizations. It is important for leaders to carefully consider the advantages and the disadvantages of internal competitive dynamics. As Dean Tjosvold, a researcher and writer on the subject, noted, “Competition stimulates, excites, and is useful in some circumstances, but those situations do not occur frequently in organizations, and the widespread use of competition cannot be justified.”


Hire, reward, and retain people in part based on their ability and willingness to work cooperatively with others for the company’s welfare Fire, demote, and punish people who act only in their individual short-term self-interest. Focus people’s attention and energy on defeating external competitive threats, not on fighting each other. Avoid compensation and performance measurement systems that create internal competition. Establish measures that assess cooperation. Build a culture that defines individual success partly by the success of the person’s peers. Encourage leaders to model the right behavior by acting collaboratively, sharing information, and helping others. Promote people to top management positions who have a history of building groups where members cooperate, share information, and provide each other mutual assistance. Use power and authority to get people and units to share information, to learn from each other, and to work collaboratively to enhance overall performance.

For media inquiries, visit the Newsroom.

Explore More