검색버튼 메뉴버튼

Why Teams Don’t Work

DBR | 1호 (2008년 1월)
An Interview with J. Richard Hackman by Diane Coutu
Over the past couple of decades, a cult has grown up around teams. Even in a society as fiercely independent as America, teams are considered almost sacrosanct. The belief that working in teams makes us more creative and productive is so widespread that when faced with a challenging new task, leaders are quick to assume that teams are the best way to get the job done.
Not so fast, says J. Richard Hackman, the Edgar Pierce Professor of Social and Organizational Psychology at Harvard University and a leading expert on teams. Hackman has spent a career exploring—and questioning—the wisdom of teams. To learn from his insights, HBR senior editor Diane Coutu interviewed Hackman in his Harvard office. In the course of their discussion, he revealed just how bad people often are at teamwork. Most of the time, his research shows, team members don’t even agree on what the team is supposed to be doing. Getting agreement is the leader’s job, and she must be willing to take great personal and professional risks to set the team’s direction. And if the leader isn’t disciplined about managing who is on the team and how it is set up, the odds are slim that a team will do a good job.
What follows is an edited version of that conversation.

You begin your book Leading Teams with a pop quiz: When people work together to build a house, will the job probably (a) get done faster, (b) take longer to finish, or (c) not get done?
That multiple choice question actually appeared on a standardized fourth-grade test in Ohio, and the obvious “answer,” of course, is supposed to be a—the work gets done faster. I love that anecdote because it illustrates how early we’re told that teamwork is good. People tend to think that teams are the democratic—and the efficient—way to get things done. I have no question that when you have a team, the possibility exists that it will generate magic, producing something extraordinary, a collective creation of previously unimagined quality or beauty. But don’t count on it. Research consistently shows that teams underperform, despite all the extra resources they have. That’s because problems with coordination and motivation typically chip away at the benefits of collaboration. And even when you have a strong and cohesive team, it’s often in competition with other teams, and that dynamic can also get in the way of real progress. So you have two strikes against you right from the start, which is one reason why having a team is often worse than having no team at all.
You’ve said that for a team to be successful, it needs to be real. What does that mean?
At the very least, it means that teams have to be bounded. It may seem silly to say this, but if you’re going to lead a team, you ought to first make sure that you know who’s on it. In our recent book Senior Leadership Teams, Ruth Wageman, Debra Nunes, James Burruss, and I collected and analyzed data on more than 120 top teams around the world. Not surprisingly, we found that almost every senior team we studied thought that it had set unambiguous boundaries. Yet when we asked members to describe their team, fewer than 10% agreed about who was on it. And these were teams of senior executives!
Often the CEO is responsible for the fuzziness of team boundaries. Fearful of seeming exclusionary—or, on the other end of the spectrum, determined to put people on the team for purely political reasons—the chief executive frequently creates a dysfunctional team. In truth, putting together a team involves some ruthless decisions about membership; not everyone who wants to be on the team should be included, and some individuals should be forced off.
We worked with a large financial services firm where the CFO wasn’t allowed on the executive committee because he was clearly a team destroyer. He was disinclined toward teamwork, he was unwilling to work at finding collective solutions, and every team he was on got into trouble. The CEO invited the CFO to stay in his role because he was a truly able executive, but he was not allowed on the senior executive team. Although there were some bruised feelings at first, in the end the CFO was much happier because he didn’t have to be in “boring” team meetings, and the team functioned much better without him. The arrangement worked because the CEO communicated extensively with the CFO both before and after every executive committee meeting. And in the CFO’s absence, the committee could become a real team.
You also say that a team needs a compelling direction. How does it get one?
There is no one right way to set a direction; the responsibility can fall to the team leader or to someone in the organization outside the team or even to the team itself in the case of partnerships or boards of directors. But however it’s done, setting a direction is emotionally demanding because it always involves the exercise of authority, and that inevitably arouses angst and ambivalence—for both the person exercising it and the people on the receiving end. Leaders who are emotionally mature are willing and able to move toward anxiety-inspiring situations as they establish a clear, challenging team direction. But in doing so, a leader sometimes encounters resistance so intense that it can place his or her job at risk.
That point was dramatically brought home to me a few years ago by a participant in an executive seminar I was teaching. I’d been talking about how leaders who set direction successfully are unafraid to assume personal responsibility for the mission of the team. I mentioned John F. Kennedy and Martin Luther King, Jr., and I got carried away and said that people who read the New Testament knew that Jesus did not convene little team meetings to decide the goals of the ministry. One of the executives in the class interrupted me and said, “Are you aware that you’ve just talked about two assassinations and a crucifixion?”
What are some common fallacies about teams?
People generally think that teams that work together harmoniously are better and more productive than teams that don’t. But in a study we conducted on symphonies, we actually found that grumpy orchestras played together slightly better than orchestras in which all the musicians were really quite happy.
That’s because the cause-and-effect is the reverse of what most people believe: When we’re productive and we’ve done something good together (and are recognized for it), we feel satisfied, not the other way around. In other words, the mood of the orchestra members after a performance says more about how well they did than the mood beforehand.
Another fallacy is that bigger teams are better than small ones because they have more resources to draw upon. A colleague and I once did some research showing that as a team gets bigger, the number of links that need to be managed among members goes up at an accelerating, almost exponential rate. It’s managing the links between members that gets teams into trouble. My rule of thumb is no double digits. In my courses, I never allow teams of more than six students. Big teams usually wind up just wasting everybody’s time. That’s why having a huge senior leadership team—say, one that includes all the CEO’s direct reports—may be worse than having no team at all.
Perhaps the most common misperception about teams, though, is that at some point team members become so comfortable and familiar with one another that they start accepting one another’s foibles, and as a result performance falls off. Except for one special type of team, I have not been able to find a shred of evidence to support that premise. There is a study that shows that R&D teams do need an influx of new talent to maintain creativity and freshness—but only at the rate of one person every three to four years. The problem almost always is not that a team gets stale but, rather, that it doesn’t have the chance to settle in.
So newness is a liability?
Absolutely. The research confirming that is incontrovertible. Consider crews flying commercial airplanes. The National Transportation Safety Board found that 73% of the incidents in its database occurred on a crew’s first day of flying together, before people had the chance to learn through experience how best to operate as a team—and 44% of those took place on a crew’s very first flight. Also, a NASA study found that fatigued crews who had a history of working together made about half as many errors as crews composed of rested pilots who had not flown together before.
So why don’t airlines stick to the same crews?
Because it isn’t efficient from a financial perspective. Financially, you get the most from your capital equipment and labor by treating each airplane and each pilot as an individual unit and then using an algorithm to maximize their utilization. That means that pilots often have to dash up and down the concourses just as passengers do, and sometimes you’ll have a pilot who will fly two or three different aircraft with two or three different crews in the course of a single day—which is not so wise if you look at the research. I once asked an operations researcher of an airline to estimate how long it would take, if he and I were assigned to work together on a trip, before we could expect to work together again. He calculated that it would be 5.6 years. Clearly, this is not good from a passenger point of view.
The counterexample, by the way, is the Strategic Air Command, or SAC, which would have delivered nuclear bombs had that become necessary during the Cold War years. SAC teams performed better than any other flight crews that we studied. They trained together as a crew, and they became superb at working together because they had to. When you’re working together in real time and there can be no mistakes, then you keep your teams together for years and years rather than constantly change their composition.

If teams need to stay together to achieve the best performance, how do you prevent them from becoming complacent?
This is where what I call a deviant comes in. Every team needs a deviant, someone who can help the team by challenging the tendency to want too much homogeneity, which can stifle creativity and learning. Deviants are the ones who stand back and say, “Well, wait a minute, why are we even doing this at all? What if we looked at the thing backwards or turned it inside out?” That’s when people say, “Oh, no, no, no, that’s ridiculous,” and so the discussion about what’s ridiculous comes up. Unlike the CFO I mentioned before, who derailed the team by shutting down discussions, the deviant opens up more ideas, and that gets you a lot more originality. In our research, we’ve looked carefully at both teams that produced something original and those that were merely average, where nothing really sparkled. It turned out that the teams with deviants outperformed teams without them. In many cases, deviant thinking is a source of great innovation.
I would add, though, that often the deviant veers from the norm at great personal cost. Deviants are the individuals who are willing to say the thing that nobody else is willing to articulate. The deviant raises people’s level of anxiety, which is a brave thing to do. When the boat is floating with the current, it really is extraordinarily courageous for somebody to stand up and say, “We’ve got to pause and probably change direction.” Nobody on the team wants to hear that, which is precisely why many team leaders crack down on deviants and try to get them to stop asking difficult questions, maybe even knock them off the team. And yet it’s when you lose the deviant that the team can become mediocre.
What makes a team effective, and how can a team’s leader make it perform better?
A good team will satisfy its internal or external clients, become stronger as a unit as time passes, and foster the learning and growth of its individual members. But even the best leader on the planet can’t make a team do well. All anyone can do is increase the likelihood that a team will be great by putting into place five conditions. (See the sidebar “How to Build a Team.”) And the leader still will have no guarantees that she will create a magical team. Teams create their own realities and control their own destinies to a greater extent, and far sooner in their existence, than most team leaders realize.
How to Build a Team
In his book Leading Teams, J. Richard Hackman sets out five basic conditions that leaders of companies and other organizations must fulfill in order to create and maintain effective teams:
1: Teams must be real. People have to know who is on the team and who is not. It’s the leader’s job to make that clear.
2: Teams need a compelling direction. Members need to know, and agree on, what they’re supposed to be doing together. Unless a leader articulates a clear direction, there is a real risk that different members will pursue different agendas.
3: Teams need enabling structures. Teams that have poorly designed tasks, the wrong number or mix of members, or fuzzy and unenforced norms of conduct invariably get into trouble.
4: Teams need a supportive organization. The organizational context—including the reward system, the human resource system, and the information system—must facilitate teamwork.
5: Teams need expert coaching. Most executive coaches focus on individual performance, which does not significantly improve teamwork. Teams need coaching as a group in team processes—especially at the beginning, midpoint, and end of a team project.

In 1990 I edited a collection of essays by colleagues who had studied teams performing diverse tasks in 27 organizations—everything from a children’s theater company to a mental-health-treatment team to a beer-sales-and-delivery team. In those studies, we found that the things that happen the first time a group meets strongly affect how the group operates throughout its entire life. Indeed, the first few minutes of the start of any social system are the most important because they establish not only where the group is going but also what the relationship will be between the team leader and the group, and what basic norms of conduct will be expected and enforced.
Off and Running: Barack Obama Jump-Starts His Team
by Michael Beschloss
If the launch of a team is as critical as Professor J. Richard Hackman says, then Barack Obama has done pretty well. He appointed his administration’s top officials much faster than most presidents do. Given the monumental crises that faced him the moment he was elected, he had to move quickly. The downside of speed was that some of his choices didn’t work out—notably Bill Richardson and Tom Daschle. Obama has certainly brought onto his team people of strong temperaments and contrasting views, starting with Hillary Clinton at the State Department and Jim Jones at the National Security Council. This suggests that we have a president who is unusually sure of his own ability to absorb differing opinions. Appointing people like Clinton also shows his eagerness to harness the talent of his former opponents. Compare that with the record of George W. Bush; his people told many job seekers who had supported John McCain in the 2000 Republican primaries, “Sorry, you backed the wrong horse!”
Of course, Obama is taking a risk by hiring so many strong and contentious personalities. He will inevitably have to spend a lot of time and energy serving as referee. This is what happened with Franklin Roosevelt, who also brought strong-minded figures into his government. One difference with Obama, however, is that FDR temperamentally loved the infighting. He liked to pit people against one another, believing that competition evoked the best performance from everyone. At times FDR actually enjoyed making his underlings suffer. I don’t think Obama does.
Most presidents prefer a happy ship, and in some cases their definition of loyalty includes not rocking the boat on major administration programs. Richard Nixon fired his interior secretary, Walter Hickel, for opposing his Vietnam War policies. There was a dissenter (what Hackman calls a deviant) on Lyndon Johnson’s team—Undersecretary of State George Ball, who strongly opposed the Vietnam War. Johnson would cite Ball when people complained that he surrounded himself with yes-men, but in fact Ball had little influence when LBJ met with top officials on Vietnam. Everyone in the group knew that Johnson didn’t take Ball’s antiwar arguments very seriously. If you really want dissenting views, better to use the Roosevelt-Obama model, where they can come from almost any member of the team—and not just from one designated rabble-rouser.
The reappointment of Bush’s defense secretary, Robert Gates, also reveals Obama’s self-confidence. He’s clearly willing to concede that there are things he doesn’t know, so he appointed someone with more than three decades of national security experience. This decision has the historical echo of John Kennedy’s near-reappointment in 1961 of Dwight Eisenhower’s defense secretary, who coincidentally was named Thomas Gates. Like Obama, Kennedy was a young president with little national security background and thought it might reassure people to have the previous defense secretary stay on at the Pentagon. Like Obama, JFK also suspected that a number of things might go wrong with national security during his first year as president. He felt that Americans might be less likely to blame the Democratic president if a Republican secretary of defense was there at his side. In the end Kennedy did not have the stomach for the risk of keeping a Republican appointee at the Pentagon. Obama did.
Obama’s first months in office prove the importance of having a president who can convey his view of the country and the world and why he thinks his plans will work. One of Hillary Clinton’s biggest criticisms a year ago was that Obama gave great speeches but that it didn’t have all that much to do with being a strong president. Obama argued that it did, and he was right. Like Roosevelt’s addresses in 1933 and Reagan’s in 1981, his public utterances—especially his speech to Congress in February—have done a lot to gain acceptance for his programs from skeptical Americans. However jaded they may be about government, Americans—even those who didn’t vote for him—are still inclined to turn to their president to explain foreign and domestic crises. Imagine how much more anxious they might feel now if Obama did not do this so effectively. Unfortunately for us all, it’s likely that he’ll have to call more on that skill as the crisis mounts in the months ahead.
Michael Beschloss has written nine books about presidential leadership, most recently Presidential Courage (Simon & Schuster, 2007).
I once asked Christopher Hogwood, the distinguished conductor for many years of the Handel and Haydn Society in Boston, how important the first rehearsal was when he served as an orchestra’s guest conductor. “What do you mean, the first rehearsal?” he asked. “All I have is the first few minutes.” He went on to explain that there’s nothing he pays greater attention to than the way he starts the first rehearsal. That’s because he knows that the orchestra members will make a very quick assessment about whether or not they’re going to make great music together, or whether he is just going to get in their way.
I do think there is one thing leaders such as Hogwood and others can do to improve the chances that a team will become something special, and that is to embrace their own quirkiness. You shouldn’t try to lead like Jeff Bezos, because you are not Jeff Bezos. Each leader brings to the task his or her own strengths and weaknesses. Exploit the daylights out of the stuff you’re great at, and get help in the areas where you’re not so good. Don’t try to ape any leadership model or team, because there’s no one right style for leading a team. There are many different ways to create the conditions for effectiveness, sustain them, and help teams take full advantage of them. The best team leaders are like jazz players, improvising constantly as they go along.
How good are companies at providing a supportive context for teams?
Perversely, the organizations with the best human resource departments often do things that are completely at odds with good team behavior. That’s because HR departments tend to put in place systems that are really good at guiding, directing, and correcting individual behavior. Take a personnel system that has been honed by industrial psychologists to identify the skills of a particular job and test individual employees on those skills. In such a system, the HR department will set up training to develop the “right” people in the “right” way. The problem is this is all about the individual. This single-minded focus on the individual employee is one of the main reasons that teams don’t do as well as they might in organizations with strong HR departments. Just look at our research on senior executive teams. We found that coaching individual team members did not do all that much to help executive teams perform better.
For the team to reap the benefits of coaching, it must focus on group processes. And timing is everything. The team leader needs to know how to run a launch meeting, so that members become oriented to and engaged with their tasks; how to help the team review at the midpoint what’s functioning well—and what isn’t—which can correct the team’s performance strategy; and how to take a few minutes when the work is finished to reflect on what went well or poorly, which can help members make better use of their knowledge and experience the next time around. Team coaching is about fostering better teamwork on the task, not about enhancing members’ social interactions or interpersonal relationships.

There’s a lot of talk about virtual teams these days. Can they work, or are they falling victim to what Jo Freeman once called the “tyranny of structurelessness”?
Virtual teams have really come into their own in the past decade, but I don’t believe they differ fundamentally from traditional teams. There was a fantasy in the beginning that everyone would be swarming around on the internet, that the wisdom of crowds would automatically prevail, and that structureless groups would come up with new and profound things that face-to-face groups could never have generated. But nirvana never materialized; virtual teams need the basic conditions for effectiveness to be in place just as much as face-to-face teams, if not more so. That said, we are seeing that we can make do with much less face-to-face contact than we ever thought possible. Today’s technology, for example, lets you have a chat window open during a web conference so you can type in the word “hand” to signal that you want to talk next. People don’t need to see your face to know that you want to speak up. But even well-structured virtual teams need to have a launch meeting with everyone present, a midpoint check-in that’s face-to-face, and a live debriefing. I don’t think for a minute that we’re going to have effective online teams if we don’t know who’s on the team or what the main work of the team really is, and so far that’s still a problem with virtual teams.
Given the difficulty of making teams work, should we be rethinking their importance in organizations?
Perhaps. Many people act as if being a team player is the ultimate measure of one’s worth, which it clearly is not. There are many things individuals can do better on their own, and they should not be penalized for it. Go back for a moment to that fourth-grade question about working together to build a house. The answer probably is that teamwork really does take longer or that the house may not get built at all. There are many cases where collaboration, particularly in truly creative endeavors, is a hindrance rather than a help. The challenge for a leader, then, is to find a balance between individual autonomy and collective action. Either extreme is bad, though we are generally more aware of the downside of individualism in organizations, and we forget that teams can be just as destructive by being so strong and controlling that individual voices and contributions and learning are lost.
In one management team we studied, for example, being a team player was so strongly valued that individuals self-censored their contributions for fear of disrupting team harmony. The team, in a spirit of cooperation and goodwill, embarked on a course of action that was bound to fail—for reasons that some members sensed but did not mention as the plans were being laid. One wonders if the crisis in the financial world today would be quite so catastrophic if more people had spoken out in their team meetings about what they knew to be wrongful practices. But again that brings us back to the hazards of courage. You’d like to think that people who do the courageous right thing and speak out will get their reward on earth as well as in heaven. But you don’t always get your reward here on earth. While it’s true that not being on a team can put your career on hold, being a real and committed team player—whether as a team leader, a deviant, or just a regular member who speaks the truth—can be dangerous business indeed.