To get important work done, most leaders organize people into teams. They believe that when people collaborate toward a common goal, great things can happen. Yet in reality, the whole is often much less than the sum of the parts.
Many teams fail because they lack the requisite experience. If you want to perform a successful cardiac surgery, you need to bring in surgeons who have mastered the techniques. If your aim is to make good stock recommendations to investors, it would be wise to hire analysts with a long track record of star performance. If your goal is to produce high-quality software, land an airplane safely, or win basketball games, you'd be smart to rely on people who have done it before. As Jim Collins put it, we need to get the right people on the bus.
But what if work experience is overrated? In a brilliant study, researchers Robert Huckman and Gary Pisano tracked more than 200 cardiac surgeons at 43 hospitals. After analyzing more than 38,000 procedures, it turned out that the surgeons didn't get better with practice. Their patient mortality rates were no better after 100 surgeries than after the first few.
A closer look at the data revealed a fascinating pattern. The surgeons did get better as they gained more experience at a particular hospital. Each procedure performed at one hospital decreased patient mortality rates by an average of 1 percent. But the benefits of experience didn't carry over to other hospitals.
The technologies weren't any different from one hospital to another; the people were. When the surgeons left their teams behind, it was as if they were starting over from scratch without any of the benefits of practice. Practice wasn't an individual act; it was a team process. As the surgeons worked with a core team of nurses and anesthesiologists at one hospital, they developed effective routines that leveraged the unique talents of each member.
In teams, it appears that shared experience matters more than individual experience. The best groups aren't necessarily the ones with the most stars, but rather the teams that have collaborated in the past. In a study of more than 1,000 security analysts led by Boris Groysberg, when star analysts moved to a new firm, it took them an average of at least five years to recover their star status -- unless they moved with their teams. The star analysts who moved alone had 5 percent odds of receiving the highest ranking from investors, whereas those who transferred with their teams enjoyed a 10 percent chance of earning the top spot.
Huckman and his colleagues found similar patterns in a study of more than 100 software development projects. The highest quality and on-time delivery rates were achieved not by the teams whose members had the most individual experience, but by the teams whose members had the most shared experience working together. Another study of product development teams showed that it typically took two to four years for members to gain sufficient experience working together to achieve their potential.
Shared experience in teams is so important that Richard Hackman, one of the world's foremost experts on teams, went so far as to include it in the very definition of team effectiveness. In Leading Teams, he argues that we in addition to assessing the quality and quantity of output, we should expand our measures of team effectiveness to include viability -- whether the team retains its capability to work together in the future.
The benefits of shared experience are visible outside knowledge work. Hackman referenced a NASA study showing that fatigued crews with experience flying together made significantly fewer errors than rested crews who had never flown together. He also pointed to an NTSB analysis of airline accidents revealing that 44 percent occurred on a crew's first flight together and 73 percent on a crew's first day. And an investigation of all NBA basketball games played from 1980 to 1994 showed that as teams gained more experience, they won more games. This was true even after accounting for player talent and age,
There are alternative explanations for some of these findings. Many airline crews only do one flight or day together, meaning that there are more chances for accidents to occur on first flights and first days. Basketball executives and coaches work harder to keep successful teams together -- and players are more motivated to stay with winning teams. Consistent with this idea, when NBA teams win more games in year 1, they're more likely to stay together in year 2. But the opposite also holds: NBA teams with more shared experience in year 1 win more games in year 2.
Interestingly, in the NBA and R&D, the gains from shared experience declined over time. The value of the first few years together was much greater than additional years accumulated. As teams stayed together longer, they had less to learn and faced a greater risk of becoming too rigid and predictable in their routines. At that point, rotating a member -- or a coach -- might be a critical step. But most teams never made it there. The vast majority of teams weren't together long enough to benefit from shared experience.
Today, too many teams are temporary: people collaborate on a single project and never work together again. Teams need the opportunity to learn about each other's capabilities and develop productive routines. So once we get the right people on the bus, let's make sure they spend some time driving together.
For more on shared experience and team effectiveness, see Adam's new book Give and Take: A Revolutionary Approach to Success, a New York Times and Wall Street Journal bestseller.