From the first time I worked in a team, I always wondered, is there an optimal team size? And how do we define the concept of the team?
A brief history of collaboration
More than 30,000 of years ago, the early modern humans (Homo Sapiens) developed the ability to think and communicate in ways that other species of humans couldn’t. Our unique language gave us a competitive edge and turned us into social animals. For the first time we could share information about the world around us but most importantly about other humans. Cooperation, coordination, and relationships became key to our survival and dominance over other species, such as the Neanderthals.
While Neanderthals hunted alone, Sapiens hunted in groups, and behaved in more flexible and adaptable ways to their environment. Although the Neanderthals were bigger and more capable when it came to one-on-one, they simply could not compete against a large group of Sapiens. Gradually, they were driven to extinction.
Working together – A competitive advantage
Our linguistic and cognitive skills allowed us to talk about other people (gossip), and understand who we could trust, developing more elaborate ways of collaboration. Although initially Homo Sapiens formed small intimate groups, we gradually started forming larger groups and tribes. However, when these social groups grew too big, they started destabilising and splitting into smaller ones.
British anthropologist Robin Dunbar suggested that there is a cognitive limit of around 150 people, with whom we can maintain stable social relationships. For a social group of this size to work people need to know each other enough to be able to gossip effectively about each other. Beyond this limit, things start breaking down. Many businesses and organisations are taking this social threshold into consideration when they organise and plan their work.
Nevertheless, humans managed to create villages, towns, cities, nations, and empires. This was enabled by the next step in our cognitive evolution, which was our ability to create fiction, legends, religions, and collective myths. These enabled us to cooperate at a much wider scale and even more flexibly with thousands or even millions of strangers.
Therefore, what gave us our competitive edge as species was our ability to think, communicate and be social. But our ability to collaborate does not scale linearly and there are cognitive limits. At a first level, instinctively, we are really good in forming small intimate groups (10 people or less). At a second level, we can use our social ability to gossip, build trust and form tribes with up to approximately 150 people and collaborate successfully, depending on certain condition. But then, it seems that we need a myth, a story or a vision to bind multiple tribes together and form a larger whole that can scale to thousands or millions.
Industrial Revolution – A renewed interest on teams
As described earlier, ever since the early humans, we have lived, worked and thrived in groups. Our social brain has given us a competitive advantage and together we managed to evolve and achieve what the first humans couldn’t even have dreamed of. Nevertheless, it was not until the industrial revolution, that we started studying teams at work.
Initially, in the late 19th century, the focus was mostly on productivity and efficiency. People were seen as cogs in a dehumanised industrial machine, where they had to do the same repetitive manual tasks day in, day out. During that time, Frederic Taylor established the concept of “scientific” management, which was focused on measuring efficiency and how fast workers could perform manual tasks. In this sense, a group of workers could be quite large, and usually bigger groups meant that the work could be completed faster. The concept of the team was not as strong as it is today, as the focus was more on the performance of a group of individuals.
Following the 1950’s a completely new era started emerging, the era of Information Technology and knowledge work. People were no longer considered cogs in a machine but were expected to think, learn, innovate, and solve complicated cognitive problems. As such, there was an increased interest in teamwork and communication.
The magical number 7 and other misconceptions
Around the same time, in 1956, cognitive psychologist George Miller published a paper called “The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information“. His paper largely examined our immediate memory and the limits in our capacity to receive, process, and store information. It also examined how information can be recoded and organised into chunks.
However, his reference to the magical number 7 is more of him being humorous rather than being scientific. Most of his paper doesn’t even focus on number 7. No wonder that in the last paragraph of his paper he wrote “What about the seven wonders of the world, the seven seas, the seven deadly sins….the seven objects in the span of attention and the seven digits in the span of immediate memory?”. He finishes saying that perhaps there is something with the number of 7 or perhaps it is a pernicious coincidence. The use of the plus or minus 2 was another humorous addition. How can a magical number have such a margin of error!
Some later research has indicated that the number 7 for memory capacity may be wrong and the real number might be closer to 4.
Nevertheless, I am not sure how this paper managed to become so popular and why people connected the number 7 to the ideal team size, despite being completely unrelated. This remains one of the biggest and more persistent myths.
Team size in sports
When it comes to sports the answer to the ideal team size seem to be quite straightforward. For instance, in football the size of the team is 11 players, in basketball it is 5 players, etc. (see table for full details). The players you see at any time on the field are unambiguously the team.
|Volleyball / Ice Hockey
|Handball / Waterpolo / Netball / Kabaddi
|Football (soccer) / American Football / Hockey / Cricket
But do these numbers represent the whole team?
The answer is no. There are substitute players sitting on the bench during the game, there is a coach and assistants, trainers, physiotherapists and various other people. All of them are part of the team, and they all contribute to the functioning of the team.
However, at any given time there is only a specific number of players on the playing field. I define them as the core team while the rest of the people that support the cause of the team are the wider team. The term core team doesn’t relate to the importance of the individuals as such but rather to the number of people that are required to be on the field and play the game at any given time. For instance in Football, the core team that is on the playing field at any given time is 11 people.
All coordination happens locally between 3 to 5 people
Nevertheless, for any particular play, there is usually a subset of players involved. For example, wherever the ball is moving, coordination usually happens between 3 to 5 people through triangular movements. Often the play might change through a long pass to a different part of the field with different players involved but the coordination is largely local. Of course, this doesn’t mean that the rest of the core team is not involved. But not everyone communicates with everyone all at the same time. The action is moving as a wave, activating different parts of the core team at any moment. Still the whole needs to work together through local action, like a complex adaptive system.
When it comes to sports with smaller team sizes, such as basketball or volleyball, usually all 5 or 6 people are involved at any given play. However, if we slow down the action we will see that most frequent collaboration at any given moment happens between 2-3 players. So, there seems to be a similar pattern.
Therefore, a team comprises of two nested parts, the core team (inner ring) and the wider team (outer ring) while at any given moment, active communication and collaboration happens only between a subset of the core team. Nevertheless, the whole team needs to be alert, sensing opportunities and positioning themselves correctly for everything to work.
In terms of team size, this largely falls between 4 and 11 people for most team sports, with the exception of rugby. However, at any given moment there is usually only 3-6 people that have to directly coordinate. This changes dynamically as the game progresses from play to play.
Team size in business
When it comes to business and knowledge work there seems to be a variety of answers in the literature. The most widely adopted framework in Agile, Scrum, suggests a team size of 3 to 9 people. Other Agile frameworks are in the same range with the only difference being the minimum number of people, which is 5. One thing is for sure though, that no individual is better than the team.
I have collected the most cited information in optimal team size in the below table.
|3 to 9
|Less than 10
|Scaled Agile Framework
|5 to 11
|Wider accepted number
(magical number 7 plus or minus two)
|5 to 9
|Research from Hackman and Vidmar
(perfect team size)
|Research from Ivan Steiner
|Elon Musk from an interview
(maximum number of people in a meeting)
|4 to 5
|Jennifer Mueller, Professor at Wharton
|no more than 5
Hackman’s research on team size
In his seminal book Leading Teams, Professor of Social and Organizational Psychology Richard Hackman describes the results of his research with Psychologist Neil Vidmar on the optimum team size. He defines as optimum team size, the one that the team is neither too small to complete the work nor too large to waste their time and effort in coordinating and engaging everyone.
For their research they created teams of 2 to 7 people and asked them to complete tasks that required decision making and judgment. They then asked them to score whether they thought the team was too small or too large. From this they found out that the “ideal” team size was 4.6.
Hackman mentions that this was based on “a not-very-important study” but the message is clear. When it comes to teams smaller is better. Yet, in his book he mentions that many organisations persist in putting together large groups of 30 or more people for political or other reasons. In such cases, he suggests the concept of the core team, which is a subset of the larger group. The core team sets the direction for the rest of the group.
Ivan Steiner’s research on group productivity
In the 1970’s psychologist Ivan Steiner did an analysis of the effect of team size on team productivity. By increasing team size from two to three people team productivity also increases substantially. But as more people are added productivity levels off and then it starts declining while process losses significantly increase due to coordination and communication issues. From Steiner’s diagrams it seems that five is the number after which productivity declines.
In 1975, Fred Brooks, a computer scientist that managed IBM’s 360 family of computers in the 1960’s, published the Mythical Man-Month: Essays on Software Engineering, offering unique insights on managing complex software projects.
In his book, Brooks talks about the mistake of using Man-Months to estimate tasks and schedule projects. He says that men and months are interchangeable only when work can be divided among workers and there is no communication required between them (i.e. workers picking cotton). However it is completely irrelevant for knowledge work, where there are constraints and complex interrelationships between tasks, requiring extensive communication.
The idea is to treat a software project as an integrated and interrelated system. In this case, communication is fundamental to make things happen. However, adding more people to a team, lengthens the schedule rather than shortens it.
Adding manpower to a later software project makes it later. The bearing of a child takes 9 months no matter how many women are assignedBrooks – They Mythical Man-Month
Communication requires a significant amount of effort, especially when there are interdependencies requiring coordination of subtasks. There is a communication channels formula n(n-1)/2, which can be used to calculate the number of communication channels.
|People in the team
The above table shows how the level of communication complexity increases with team size. As such, the smaller the team the less communication effort there is.
Less communication is better
To take this a step further, Jeff Sutherland, the co-creator of Scrum, writes that less communication is better, as communication doesn’t scale. In traditional project management, the thinking was that by adding extra people the project will finish earlier but as we saw this is just a myth. Instead, Jeff Sutherland suggests that it is better to have many small, decoupled, self-managing teams rather than one large team. Instead of adding people, it is better to add more of these teams and spread the workload without creating additional communication overhead.
The Ringelmann effect
In the early 1900s a French engineer, Maximilien Ringelmann, asked several teams of people of different sizes to compete in pulling a rope. He discovered that individual productivity diminishes as the team size increases. In other words, people didn’t put as much effort the bigger the team was. In social psychology this phenomenon, where people contribute less in the effort of the team, is described as Social Loafing. For some reason, people hide inside the group, feeling less accountable for the success of the team, and expecting others to compensate for their part.
Therefore, what the Ringelmann effect shows us is that the smaller the team, the more engaged, responsible, and accountable the team members will feel.
It’s a kind impulse, but too-large groups usually backfire. “If you don’t break it down, people will break it down for you. They’ll form cliques”.Jennifer Mueller, Professor at Wharton University
What is the optimal team size
In a nutshell, most academics, researchers, and practitioners agree approximately that teams that do complex interdependent work need to be small (ideally no more than 5 people). This keeps communication overhead to a minimum (no more than 10 channels) while making sure that everyone in the team contributes to the maximum. However, for this to work, a team needs to have a minimum of 3 people, or else they may struggle to complete the work. Therefore, based on my research I believe that the ideal team size should be between 3 and 5, with a preference for 5.
As mentioned, in team sports with bigger team sizes, i.e. football (11 players), what naturally happens is that a subset of the team is actively involved at any given moment as the action moves across the field. This doesn’t mean that the rest of the team is not paying attention or positioning based on what is going on. But rather that at any given moment the action, and line of sight communication and coordination takes place between 3 to 5 people, who pass the ball between each other.
On the other hand, basketball seems to have the idea number of players (five), which is just enough for continuous flow of communication and effort, while avoiding the Ringelmann effect.
As Homo Sapiens, we have come a long way since our cognitive revolution, more than 30,000 years ago. Our ability to communicate, coordinate, and collaborate in small, intimate social groups, and then in larger tribes, has given us a distinct survival advantage.
However, there are some communication thresholds, over which our ability to communicate and work together diminishes significantly. To overcome the communication threshold of having too many individuals working interdependently, my proposed solution is to build core teams that are quite small in size (no more than five people). To overcome the second threshold of having too many teams working interdependently my proposed solution is to create a compelling vision that binds them all together into a loosely-coupled team of teams.
After all, when it comes to ideal team size, there might not be a magical number seven, but number five seems quite magical to me.