Why Less Is More in Teams
Why is it that American football uses eleven players, Canadian football twelve, and Gaelic fifteen? Why eleven in European soccer? Why does baseball field nine players, basketball five, volleyball six, water polo seven, and cricket eleven? Simple as these questions are, they are deceptively hard to answer. But whatever the historical reasons, the number of people on a team has significant impact on performance. Here’s why.
The earliest known attempt to investigate the relation between team size and productivity dates back about a hundred years to the now famous experiments by French engineer, Maximilien Ringelmann. In a set of simple rope pulling experiments he discovered that, in what is now known as the Ringelmann Effect, people’s efforts quickly diminish as team size increases. Eight people, he found, didn’t even pull as hard as four individuals. He rationalized the decay in effort by suggesting it was difficult for team members to coordinate effort, and left it at that.
In a brilliant twist on Ringelmann, Alan Ingham and three colleagues in the 1970s decided to recreate the experiment in the basement of the University of Massachusetts Amherst. Whereas their replica of Ringelmann’s experiment produced near-identical results, a clever variation on it — where, unbeknownst to others, some students were asked to merely pretend they were pulling the rope — generated an extraordinary observation. It didn’t seem to matter whether people were part of a larger team or simply thought they were part of a larger team — they worked less hard. Thus in a team of six (where three had been asked to pull as hard as they could and three others instructed to pretend), those actually pulling the rope put in only as much effort as they had previously done in teams of six.
Ingham and his colleagues had demonstrated that loss of effort could not be explained by lack of coordination, as Ringelmann originally thought. Their experiments instead illustrated the problem ofsocial loafing — when team members reduce their effort because they feel less responsible for the output.
Social loafing is one of the most documented phenomena in social psychology, and has been demonstrated on all kinds of teams including those that rely on people with different skill sets working in some coordinated fashion (such as those in today’s workplace, since team tasks such as pulling a rope are relatively rare in the workplace). Since Ingham’s recreation of Ringelmann’s experiment, at least another eighty studies on social loafing have been published, based on a variety of tasks — including such complex tasks as brainstorming or rating poems for quality. In these experimental contexts, the research shows that people tend to prefer teams of four or, at most, five members. Anything lower than four was felt to be too small to be effective, whereas teams larger than five became ineffective.
The difficulty with studies of this type — and their applicability to today’s workplace — is that they do not always account for task variety, where some tasks require a much larger skill set than represented by four or five individuals. Nor do they account for some people being more skilled at managing large teams than others. But they do remind us that smaller teams are generally better and, all other things being equal, that teams are more likely to optimize their performance when faced with slightly fewer members than the task at hand requires.
If reducing the size of your current team is not an option, how might you go about preventing or tackling social loafing? Not doing anything isn’t an option — the team members you risk losing aren’t the weakest but highest performers. Your best performers typically resent the company of those who don’t pull their weight, particularly if the reward system doesn’t adequately discriminate between average and top performance.
One option is to divide up a complex task into manageable bits, where every member of the team is accountable for one bit of it. Purists will argue that this no longer constitutes a team but a working group, with one central figure to coordinate everyone’s output.
A second option is to generate a sense of urgency. As I’ve pointed out elsewhere, the key to effective teamwork is surprisingly straightforward: Provided that people are capable, all one needs to do is to give them something to care about more than themselves. The real problem is that this is difficult to do consistently.
A third option is to make weaker team members feel disproportionately responsible for the team underperforming — something formally known as the Köhler effect and, as far as I’m concerned, unappetizing and Machiavellian.
A fourth option is to provide greater transparency by opening up your feedback mechanism. Sports have the edge here in that, particularly in elite environments, systems are designed to measure performance on an ongoing basis, aided by increasingly sophisticated technologies. For example, in early 2007 the Cambridge University Boat Club embarked on a controversial experiment. Each week they would post not only objective performance results on the inside door of their ground floor gym but, right next to it, a sheet with five columns to cover subjective assessments by the coaching team. With your name in the first column, the coaches told you what you did well in the second column, and what you didn’t do so well in the third column. The fourth and fifth columns were reserved for what you should stop doing, or start doing, immediately. Though somewhat uncomfortable, this left few places to hide.
Reality is such that few corporate environments allow for that sort of transparency. The question is what will?