The Failure-Tolerant Leader
Originally published in the Harvard Business Review, the article won the first-place McKinsey Award for 2002. The McKinsey Awards honor the best articles published in HBR every year, as determined by an independent panel of judges.

Executives know that failure is an integral part of innovation. But how do they
encourage the right kinds of mistakes?

By Richard Farson and Ralph Keyes

Richard Farson is the cofounder and president of the Western Behavioral Sciences Institute in La Jolla, California. He is the author of Management of the Absurd: Paradoxes in Leadership (Simon & Schuster, 1996). Ralph Keyes is the author of Chancing It: Why We Take Risks (Little, Brown, 1985) and The Courage to Write: How Writers Transcend Fear (Henry Holt, 1995). This article is adapted from Whoever Makes the Most Mistakes Wins: The Paradox of Innovation, by Richard Farson and Ralph Keyes. Copyright 2002 by Richard Farson and Ralph Keyes. Reprinted by permission of the Free Press, a division of Simon & Schuster.

"The fastest way to succeed," IBM’s Thomas Watson, Sr., once said, "is to double your failure rate." In recent years, more and more executives have embraced this point of view, coming to understand what innovators have always known: that failure is a prerequisite to invention. A business can’t develop a breakthrough product or process if it’s not willing to encourage risk taking and learn from subsequent mistakes.

The growing acceptance of failure is changing the way companies approach innovation. Some build exit strategies into their projects to ensure that doomed efforts don’t drag on indefinitely. Others, like the credit card company Capital One, continually conduct large numbers of market experiments knowing that while most of their tests won’t pay off, even the failures will provide valuable insights into customer preferences. Still others launch two or more projects with the same goal, sending teams in different directions simultaneously. This approach—called "simultaneous management" by civil engineering professor Alexander Laufer—creates the potential for a healthy cross-fertilization of ideas and techniques.

While companies are beginning to accept the importance of failure in the abstract—at the level of corporate policies, processes, and practices—it’s an entirely different matter at the personal level. Everyone hates to fail. We assume, rationally or not, that we’ll suffer embarrassment and a loss of esteem and stature. And nowhere is the fear of failure more intense and debilitating than in the competitive world of business, where a mistake can mean losing a promotion, a bonus, or even a job.

During his years leading Monsanto, Robert Shapiro was struck by how terrified his employees were of failing. They had been trained to see an unsuccessful product or project as a personal rebuke. Shapiro tried hard to change that perception, knowing that it hindered the kind of creative thinking that fueled his business. He explained to his employees that every product and project was an experiment and that its backers failed only if their experiment was a halfhearted, careless effort with poor results. But a deliberate, well-thought-out effort that didn’t succeed was not only excusable but desirable.

Such an approach to mistake-making is characteristic of leaders we call "failure-tolerant executives who, through their words and actions, help people overcome their fear of failure and, in the process, create a culture of intelligent risk taking that leads to sustained innovation. These leaders don’t just accept failure; they encourage it. We’ve studied a number of failure-tolerant leaders -- in business, politics, sports, and science -- and found some common threads in what they do. They try to break down the social and bureaucratic barriers that separate them from their followers. They engage at a personal level with the people they lead. They avoid giving either praise or criticism, preferring to take a nonjudgmental, analytical posture as they interact with staff. They openly admit their own mistakes rather than covering them up or shifting the blame. And they try to root out the destructive competitiveness built into most organizations.

First and foremost, though, failure-tolerant leaders push people to see beyond simplistic, traditional definitions of failure. They know that as long as someone views failure as the opposite of success, rather than its complement, that person will never be able to take the risks necessary for innovation.

Move Beyond Success and Failure

Of course, there are failures and there are failures. Some mistakes are lethal – producing and marketing a dysfunctional car tire, for example. At no time can management be casual about issues of health and safety. But encouraging failure doesn’t mean abandoning supervision, quality control, or respect for sound practices. Just the opposite. Managing for failure requires executives to be more engaged, not less. Although mistakes are inevitable when launching innovation initiatives, management cannot abdicate its responsibility to assess the nature of the failures. Some are excusable errors; others are simply the result of sloppiness. Those who are willing to take a close look at what happened and why, can usually tell the difference. Failure-tolerant leaders identify excusable mistakes and approach them as outcomes to be examined, understood, and built upon. They often ask simple but illuminating questions when a project falls short of its goals:

Distinguishing between excusable and inexcusable failure offers two broad benefits. First, it gives managers a tool to build a nonpunitive environment for mistake making while allowing them to encourage thoughtfully pursued projects that, should they fail, will yield productive mistakes. Second, it allows managers to nonjudgmentally promote the sort of productive mistake making that is the basis for learning. By revealing what doesn't work--in the lab or in the marketplace--a failure flowing from a carefully designed and executed project provides insight into what will work.

Success can be approached in much the same way. Like mistakes, all successes are not created equal. A success due to a fortunate accident is not as desirable as one resulting from a thoughtfully pursued project. Embracing them as equals would send a message that the end justifies the means -- that success achieved through poor work is desirable. Thus, successes might be evaluated with questions similar to those posed about failures. How much was due to good fortune, how much to the hard work of its creators? Were all contributors acknowledged? Did the success move us closer to our goals? Will it actually serve customers’ needs or simply merit an award from peers? By taking this perspective and raising such questions, managers can begin to treat success and failure similarly, more like the siblings they actually are.

Some managers may find that idea difficult to embrace. Treat success and failure the same? Shouldn’t I reward success? And even if I don’t reprimand an employee who fails, shouldn’t I at least call attention to the mistake? Well, no. We suggest a different approach.

Get Engaged

The best coaches take victory and defeat in stride. "I didn’t get consumed by losses," said the legendary NFL coach Don Shula, "and I didn’t get overwhelmed by successes." Failure-tolerant leaders do the same. Rather than pursue success, they focus on increasing their organizations’ intellectual capital: the experience, knowledge, and creativity of the workforce. How? Through engagement. These managers take a tangible interest in their employees’ projects. Instead of simply evaluating an employee’s efforts, they try to understand the work, interpret it, and discover its meaning to the individual. Often, they are in a position to see the work in a larger context, making them the ideal people to discuss a project’s history, goals, and larger significance to the organization.

That process is more collaborative than supervisory. Failure-tolerant managers show interest, express support, and ask pertinent questions: What’s new with your project? What kinds of problems are you having? Taking the long view, what might the next steps be? Conversations are less about whether the project is succeeding or failing than about what can be learned from the experience. When a manager and employee are deeply engaged in that discussion, both enter the same kind of high-performance zone that athletes do when they’re at their very best. In this zone, evaluation is less relevant than the subject of where to go from here.

Enlightened managers strive to be collaborative rather than controlling. Only through engaged conversations over time can managers create failure-tolerant work environments that invite innovation. This is not to say that a major achievement shouldn’t be applauded, or that repeated, avoidable mistakes should be tolerated. But astute managers mark the daily progress of small successes and failures with an evenhanded, open curiosity about the lessons learned and the next steps to take.

Listening is more central to this process than talking. Research on workplace creativity shows that it’s not the individual employee’s freedom as much as the managerial involvement that produces creative acts. No incentive can match the obvious appreciation shown by a manager’s interest and enthusiasm. Path-breaking leaders such as Thomas Edison, General Motors’ Charles Kettering, the Watsons at IBM, and 3M's William McKnight were famous for schmoozing with employees—not second-guessing or criticizing, but engaging in animated discussions about projects. Nothing does more for productivity, morale, and employee retention. "Edison made work interesting," said a machinist and draftsman who spent a half-century working for the inventor. "He made me feel that I was making something with him. I wasn’t just a workman."

Don’t Praise, Analyze

New ideas are most likely to emerge in the workplace when managers treat steps in the innovation process—those that work and those that don't—with less evaluation and more interpretation. They don’t praise or penalize; they analyze.

Less praise? Haven’t managers been told not to skimp on compliments? They have indeed. But psychologists who have studied the effects of praise question its value. As with criticism, compliments can actually demotivate people. Recipients may feel manipulated or think that too much is expected of them. Research has found that children playing games lose interest once they're rewarded or complimented for their play. In one study, students praised less by their science teachers did a better job of conducting experiments on their own than the ones who were praised more. That is why, in place of perfunctory praise, many educators are shifting to a teaching style in which they ask questions, give feedback, and show interest, but are spare with compliments. "That’s great!" gives way to "I see you’ve decided to use liquid nitrogen in this experiment." Such a specific response shows real interest in a student’s work which is appreciated more than repeated praise is. In the workplace, praise can become what is called a "dissatisfier." Like a salary, it is less likely to motivate when it’s given out than demotivate when it’s expected but withheld. So a manager cannot suddenly stop praising an employee who has come to expect it. But when an engaged manager takes a genuine interest in an employee’s work, the need for compliments declines.

Genuine engagement, though, can require far more time than the 11 minutes per task that managers spend, on average. Because involvement takes more time than keeping your distance, occasions for doing so must be chosen carefully. Engaging with employees is demanding and risky; it can threaten a manager’s authority. The more involved you get with employees, the harder it becomes to reprimand them when necessary. Although not the same as personal friendships, engaged professional relationships resemble them in ways that can hinder the supervisory process. The challenge is to learn how to get closely involved with an employee’s work without presuming to be pals.

Managers may be wary of this type of engagement because it can be unpredictable, raising questions they can’t answer or might rather avoid. But that’s a price worth paying. The open-ended, less formal nature of an engaged relationship can lead to the unexplored terrain where innovation lies.

Earn Empathy

While the notion of encouraging mistakes may seem alien—or at least a little unnerving —to many managers, it has some celebrated champions. When Jack Welch was head of GE, he said, "We reward failure," explaining that to do otherwise would only squelch daring. GM’s Charles Kettering, who was regarded as second only to Thomas Edison as America’s leading inventor mogul, liked to say that a good research man failed every time but the last one. "He treats his failures as practice shots," Kettering noted, adding that he himself had been wrong 99.9% of the time. What every educated person needed to learn, he felt, was "that it's not a disgrace to fail, and that you must analyze each failure to find its cause….You must learn how to fail intelligently. Failing is one of the greatest arts in the world. One fails forward toward success."

Welch and Kettering knew that creating a risk-friendly environment requires demonstrating unequivocally – in deeds more than words – that stumbles on the innovation path are forgiven. How better for managers to achieve this end than by publicizing their own missteps? The late Roberto Goizueta got years of one-liners from the New Coke fiasco that he sponsored. Admitting his mistake conveyed to his employees better than a hundred speeches or a thousand memos that "learning failures," even on a grand scale, were tolerated.

A former Lockheed executive recalled the time CEO Dan Houghton gathered his company's manufacturing heads to discuss his own errors. One notable mistake, Houghton recounted, was when Howard Hughes once called to tell him that Douglas Aircraft was in trouble and that he should get right over there and buy it "before Jimmy McDonnell [did]." Houghton told his managers that not taking Hughes's advice was the biggest blunder he had made in business. Had he bought Douglas Aircraft, Houghton could have consolidated his company’s operations in Southern California and avoided costly future moves.

The former Lockheed executive said that he and his colleagues left the meeting with increased respect for Houghton and renewed motivation to take their own risks. Houghton had told them in the most eloquent way possible that losing a gamble wasn't the worst thing that could happen to them. The message they took away: "If Dan Houghton can make mistakes, I guess I can, too."

Similarly, managers at 3M reinforce the company’s mistake-tolerant atmosphere by freely admitting their own goofs. Former CEO L.D. DeSimone never hesitated to recount how he repeatedly tried to stop the development of Thinsulate. Luckily, DeSimone failed, and Thinsulate became one of the company’s most successful products. By being so candid about his near blunder, DeSimone powerfully conveyed that it's okay to be wrong and to admit it when you are.

Far from revealing weakness, admitting mistakes shows a leader’s self-confidence. It helps forge closer ties with employees and colleagues. A blunder admitted is empathy earned. Leaders who don’t cover up their errors reveal themselves as human—they become people whom others can admire and identify with.

Think of Ulysses S. Grant, whose success as a military officer grew directly out of his earlier problems in civilian life. Grant’s well-known battles with alcoholism made him more sympathetic to others’ weaknesses; his struggle to control his addiction helped him understand and deal with lapses of discipline among subordinates. Grant had also failed repeatedly in business, suffering humiliating experiences that made him more humble than other officers. His failures also made him more daring. With little to lose, Grant took more chances during combat than did a general such as George McClellan, who had enjoyed a brilliantly successful life before the war, but proved too cautious by far on the battlefield.

The greatness of many U.S. presidents was also built on a foundation of unremarkable performance early on. Before Valley Forge, Washington failed repeatedly as a military officer and did a so-so job of running his plantation. Lincoln suffered multiple setbacks in business and politics. Franklin Roosevelt—a mama's boy and a mediocre student—was generally considered a political lightweight before he moved into the White House. It's taken for granted by historians that if he hadn't been paralyzed as a young man, Roosevelt would never have developed the grit, depth, and empathy he needed to become one of our best presidents. FDR's successor, Harry Truman—today ranked among our top half-dozen chief executives—didn’t go to college, was a mediocre farmer, and failed as a haberdasher. "Most of our better presidents," observed Ronald Reagan’s campaign director John Sears, "learned to empathize through suffering personal tragedy or failure. … There is something about losing and coming back from it that burns character into a man's soul, breeds confidence without arrogance, and makes a man believable when he talks about problems."

Employees do better when they know they're being supervised by a human being, not a detached director. This doesn't mean they spurn self-confident leaders. We all want to believe that our leaders are competent and have exceptional qualities. But we sometimes want a bit of the opposite as well. An inscrutable mask makes it harder to lead. Both vulnerability and transparency are important for leaders and organizations alike.

Collaborate to Innovate

Creating a culture in which employees feel comfortable with failure also requires abandoning traditional ideas about personal competition. The idea that achievement is maximized when we go at one another tooth and nail is engraved on our national psyche. But when the road to success requires making others fail, innovation gets left by the wayside. Competition infects coworkers with a desire to win rather than to solve problems and move projects forward. In the process, employees inhibit the free flow of information that’s so vital to innovation. Those who feel their work is being judged on conventional concepts of success and failure, and who feel they’re competing with coworkers for the brass ring, will want to protect information rather than share it. This is a textbook way to squelch innovation. Companies become the losers. If 3M’s Spencer Silver had thought it prudent to conceal information about the imperfect adhesive he invented with another colleague, Post-It Notes might never have been invented.

Prizes for performance are especially effective at undermining teamwork because they place competition above collaboration. A food services company we'll call Comestibles once had a contest that awarded regional offices that posted the best sales records. Those who won got free vacations. This competition produced a few happy winners and lots of disgruntled losers. Winning became such a fetish that Comestibles’ employees began hoarding information they might otherwise have shared with one another. Some even fudged their figures to gain an edge. Rather than encourage employees to collaborate and share information, Comestibles' motivational program created an atmosphere of competition that stifled creativity, openness, and honesty.

Meanwhile, some future-minded companies like Shell and Monsanto have developed work groups that emphasize collaboration. The main objective of these groups is to exchange information, not hide it, as so often happens in the heat of competition. 3M has encouraged idea sharing for decades, from the coffee-and-donut skull sessions years ago (like the one where Spencer Silver discussed his not-sticky-enough glue) to today's more formal Tech Forums and in-house trade shows.

Failure-tolerant leaders understand that collaboration is the real road to innovation. They see it as the best means for tapping into the imaginations of employees who are not especially competitive but who might have invaluable, innovative ideas. Because such people don’t feel the need to win every exchange of ideas, they don't do well in gatherings of colleagues playing verbal king-of-the-mountain. Competition-based cultures can be especially hard on new hires, introverts, minorities, women, loners, and those for whom English is a second language. It's difficult to exaggerate the stifling effect that competitive idea exchange can have on such people. That’s where communication technologies can help.

Communications technologies have opened the doors for idea sharing. In organizations of all kinds, they have encouraged bottom-up decision making, loosened sticky spigots of information, and become an electronic suggestion box. Royal Dutch/Shell, for instance, has found that many of its most worthwhile innovations have come from employees via e-mail. Similarly, IBM's rebirth as a more nimble, Web-savvy company began when CEO Lou Gerstner invited employees at all levels to communicate with him by e-mail soon after he arrived in 1993. It did not take long for Gerstner to begin hearing about stalled projects and unsolved problems that were stuck in IBM's many corporate cobwebs.

E-mail, of course, is only the beginning. Communications technologies now extend well beyond basic e-mail into various kinds of both real-time and asynchronous electronic connections. They include chat rooms and news groups; conferencing systems; technologies for surveys, voting, and joint document preparation; distance education; remote control of computers for demonstrations or group graphic designs; and more. In Robert Shapiro’s words, many of these technological developments "just rip through hierarchy."

Electronic communications are ideal for involving creative people who might be shunned or perceived as marginal in organizations that rely too heavily on face-to-face idea exchange. With many of these technologies, an employee's age, gender, ethnicity, physical appearance, or personality quirks no longer determine how his ideas are received. Moreover, because these forms of communications don’t always require face-to-face meetings, employees have the time to prepare and edit their proposals. These benefits have made it easier for companies to retain unconventional, creative employees who may be bad at office politics but good at generating fresh ideas.

Give a Green Light

At Royal Dutch/Shell, innovation teams fielded suggestions e-mailed by fellow employees. Using a method devised by the consulting firm Strategos, these six-person "GameChanger" teams met weekly to assess ideas. As Gary Hamel, the head of Strategos, reported in his article "Bringing Silicon Valley Inside" (HBR September 1999), the GameChanger teams assessed 320 proposals during their first two years of operation. The proposals were evaluated not just on the basis of what Shell stood to lose by pursuing the suggestion, but what it could lose by not doing so. Of Shell's five top innovations in 1999, four bubbled up through GameChanger meetings. One of these was the company’s Light Touch method of using laser sensors to discover hydrocarbon emissions of oil deposits. Among the most striking discoveries of the GameChanger process was how many ideas came from employees who weren’t thought to be innovative. It turned out that most had never had opportunities to express their ideas.

Enlightened leaders emphasize that a good idea is a good idea, whether it comes from Peter Drucker, the Reader’s Digest, or an obnoxious coworker. This approach blunts the group's natural disposition to squelch imaginative, though difficult, participants. Psychologist Michael Kahn suggests running meetings using the "barn raising" model, based on the way pioneers pitched in as a community to help one another construct outbuildings. According to this model, rather than engage in one-upmanship, members are encouraged to listen carefully to each person’s idea, then add their thoughts to see if they can build that idea into a valuable contribution. Such an atmosphere of exploration lets group members search diligently for value in ideas that might otherwise have been discarded. They also feel comfortable knowing that their suggestions will receive the same treatment.

Like Shell’s GameChanger groups, Kahn’s barn-raising techniques are an effective catalyst for innovation. So, too, is a gathering known as a "community of practice," or a small group of employees within an organization that meets regularly to discuss common interests. The value of these communities goes far beyond information exchange. In their article "Communities of Practice: The Organizational Frontier" (HBR, January-February 2000), management consultants Etienne Wenger and William Snyder refer to these gatherings as "petri dishes for entrepreneurial insights." The way these groups arrive at ideas is similar to brainstorming: The members suspend judgment and allow others to toss out suggestions in an atmosphere of cooperation and support rather than competition and criticism. Unlike skunk works, these groups are ad hoc gatherings of people from many parts of an organization; their collective attitudes can often infuse an organization's culture as a whole, particularly an attitude of daring. As the member of one Indiana community of practice told Wenger and Snyder, "I took a risk because I was confident I had the backing of my community -- and it paid off."

What’s really going on in these groups is courage enhancement. By creating an atmosphere of safety and reducing the pressure to succeed, the groups give people the confidence to share their ideas. Employees who once felt inhibited suddenly feel free to express their thoughts, frequently contributing to the innovations that drive the company. In the process they re-define their concept of "success" and "failure."

Once conventional notions of success and failure are discarded, many more management approaches become possible. A both / and rather than either / or approach makes it easier to promote innovation by encouraging risk-taking, supporting mavericks, and doing simultaneous planning. At the heart of this management posture is greater acceptance of failure as a necessary part of the innovation process.

Leaders who convey this send a clear message to their organizations that constructive mistakes are not only acceptable but invaluable. Employees feel that they have been given a green light to explore new approaches, think less about success or failure and more about learning and experience. That is the key to coming up with breakthrough products and processes: viewing mistakes as brightly colored signposts on the road to success.