Large teams don’t work
When it comes to team size, less really is more
The promise of teamwork
In today’s business world, the vast majority of work is done by teams. Surveys estimate that 65% to 95% of knowledge workers in the United States and Europe are part of more than one team at a time. In fact, our Western corporate culture places a massive emphasis on the importance of working together. Nowadays, you would be hard-pressed to find a vacancy that doesn’t mention vague buzzwords such as ‘being a team player or ‘having great communication skills’. It is almost unthinkable that you would have a job in which you would not work together with others at all — be they your permanent colleagues or some temporary project partners. If someone shows a preference for working alone, we often perceive them to be socially awkward or even arrogant. From a young age, children learn to ‘play well together’ and to work in small teams on various projects in school. All of this begs the question: is working together always better than working alone? While that is a complicated question, the short answer is ‘nope’.
Don’t get me wrong — I am not bashing teamwork in general. Quite the contrary: working together allows us to achieve better outcomes than we would be able to achieve if we worked alone. That is because, in theory, teams have two substantial benefits over individuals: firstly, they have access to a broader pool of knowledge, expertise, and experience, and secondly, they are better able to exercise oversight. That first advantage is due to a process called information elaboration. Without getting too much into the technicalities of information processing in teams, you can compare information elaboration to solving a jigsaw puzzle when the pieces are distributed over multiple people. In order to complete the puzzle, you need to share your pieces with the rest and hope they share their pieces with you as well. The second advantage, being able to exercise oversight, must ring a bell: as a team member, your choices are in a way ‘checked and controlled’ by the other team members. If you make a mistake, others have a chance to correct your mistake before it’s too late. Taken together, information elaboration and effective oversight allow teams to make the right decisions in complex situations. In theory, at least. That’s why we leave our most complex decisions to teams. Parliament? A team. The board of directors? Also a team. A jury? That’s a team as well. However, as you might expect, the reality is not as straightforward. Teamwork rarely makes the dream work.
What we often experience in the real world is that working together well is hard. It takes careful planning, working both hard and smart, and continuous reflection. These problems are already challenging enough in small teams, but become even larger as the team size increases. That has many causes, and some of those causes are explained particularly well by a rather obscure observation: Price’s square root law.
What is Price’s law?
Derek de Solla Price, a British scientist and historian, observed the following: in any area of scientific literature, half the publications come from the square root of all contributors. In more general terms: 50% of the outputs are created by the square root of all participants. This effect can be observed across many fields. In literature, half of the books that are sold are written by a very small percentage of authors. In music, half of the album sold are composed by a very small percentage of musicians. And in your average company, half of the sales are generated by a very small percentage of salespeople.
Let’s look at an example. Let’s assume your company has a sales department, in which 100 employees work. Let’s assume that this sales department makes about 500 successful sales every week. Theoretically, we could expect that every salesperson makes about 5 successful sales per week: 500 divided by 100 equals 5. Even if the distribution is not as uniform like that, the differences between the best and the worst performers are not too drastic, right? According to Price’s law, however, that is most likely not the case. Instead, half of the sales are made by the square root of the number of salespeople. This amounts to 250 sales being made by the 10 most productive salespeople, with the other 250 sales being made by the 90 least productive salespeople. That is a massive difference in productivity! Think of the teams you worked in in the past. Was the work distributed evenly amongst all team members? Did the fairness of that distribution change as the team increased in size?
An important implication of Price’s law is that the larger a team becomes, the larger the proportion of unproductive team members becomes. We can illustrate this with another example. Let’s assume a team of 4 people. The square root of 4 is 2, so according to Price’s law, 2 people do half of the work. That means that the work is split roughly equally over all team members. Let’s increase the size of that team to 25. The square root of 25 is 5, so Price’s law states that only 5 people do half of the work. While the total team size increased by 21 members, the amount of highly productive team members increased only by 3. We can draw the conclusion that as team size increases, competence scales linearly and incompetence scales exponentially.
It is important to understand that Price’s law does not predict the total amount of work being done. For obvious reasons, a team of 25 can get more work done than a team of 4. However, increasing team size increases the differences in productivity between team members. The differences between the most and the least productive team members become more pronounced.
Why does Price’s law occur?
It is very difficult to properly conduct an experiment to probe for the root causes of Price’s law. However, decades of research into human decision science and behavioral economics have helped us understand some very important theoretical concepts, which we can use to explain Price’s law. One such concept is called ‘social loafing’. Social loafing refers to the fact that people tend to work less hard in a larger team than they would in a smaller team or individually. Under the umbrella of social loafing fall many psychological phenomena. One of those is the diffusion of responsibility. This concept is often related to the bystander effect. Imagine walking down the street and seeing a child drowning. Would you jump into the water to save the child? We all like to imagine ourselves being the hero in this story, but psychology tells us that we should not be so sure of that. The bystander effect states that the more people are present to witness a crime or a disaster, the less likely we are to help the victim. When push comes to shove, why should we be the first to help? Why can’t other people help? This example perfectly explains the diffusion of responsibility. It is also exactly what happens in large teams: why should we work hard if we share responsibility with many others? After all, it would be next to impossible to focus exclusively on our efforts. Large teams allow non-performers to slack off safely.
A second relevant concept within the domain of social loafing is the dispensability of effort. This also leads to people slacking off, but not because they feel they can ‘afford’ slacking — rather because they feel that their efforts won’t matter much in the grand scheme of things. It is an argument that is often used by people who do not vote in elections. Essentially, one single vote is just another drop in the ocean. In larger teams, individual contributions feel less significant, which causes team members to be less motivated and therefore work less hard.
To make matters even worse for managers: bad apples spoil the bunch. In a way, social loafing is contagious. That is due to two effects: the sucker effect and downward matching. The sucker effect might feel familiar. You experience it whenever you feel that other team members might want to leave you to do all the work. Because you don’t want to be a ‘sucker’, you wait and see how much effort the other team members are willing to put into the project. Only if you see that they are serious about the project are you willing to work hard as well. Downward matching has everything to do with our innate desire for justice and equity. If we feel like we are doing much more work than the other team members, we are inclined to reduce the amount of effort we exert. After all, it wouldn’t be fair if we had to do all the work!
Practical implications for managers
I must stress that, even though Price’s law is called a ‘law’, it does not universally hold true. There are definitely situations in which outputs are spread more equally over participants. I would therefore argue that is not relevant for us to debate the exact limitations of Price’s law. Instead, I propose that leaders use Price’s law as a starting point for discussion — as a tool that enables them to strategize better and to make more effective decisions — instead of as a rigid lens through which to see the world. Seen this way, Price’s law has several practical implications for managers, especially for those managing knowledge workers and creative teams. Some of these implications are the following:
- Keep creative teams small
Remember that competence scales linearly, while incompetence scales exponentially. The larger your team is, the larger the proportion of relatively unproductive members will be. A number that is often quoted is that the optimal amount of members in a creative team is three. According to Menno van Dijk and Robert Wolfe, “anything smaller lacks the diversity to generate creative perspectives, and will seem under-resourced, while larger teams suffer from the complexity of coordination.” Small teams seem to outperform large teams on criteria such as creativity, innovativeness, and efficiency. - Be slow to hire
According to Price, the added (or marginal) productivity of each extra team member becomes smaller and smaller the more team members you recruit. This could lead to a situation in which certain team members actually hurt your bottom line: their productivity is lower than what they are costing you. Ideally, team size should only be increased when it is highly likely that an extra team member would increase productivity exponentially. - Adjust your compensation strategy to retain top performers
Many traditional companies use a deeply flawed compensation strategy. Most workers are paid more or less the same because it is assumed that the differences in productivity between employees are relatively small. According to Price’s law, this is not the case. The most productive team members are exponentially more productive than the least productive team members, which makes them much more valuable. In order to retain them and their productivity, they should be compensated fairly. Furthermore, rewarding individual performance discourages social loafing. However, performance-based pay is not without its problems — if it is implemented wrongly, it can actually hurt team performance more than help it. So, proceed with caution. - Increase interdependence to discourage social loafing
We have seen that social loafing is likely an important root cause for Price’s law to occur. Therefore, we want to discourage social loafing as much as possible. We can do so by keeping teams small, of course. But another effective way of decreasing social loafing is by increasing interdependence in the team. Interdependence refers to the extent that team members need each other in order to achieve the team’s goals. If everyone is indispensable, individual team members will keep each other to high standards. They will be motivated to work hard and work well together in order to be successful.
