Reading Time: 4 minutes

Communities of practice have a significant impact on organisations for all the reasons which I have spoken about in my book and many times on stage. I have been fortunate to have the opportunity to explore the correlation between communities of practice and natural human communities with evolutionary psychologist Robin Dunbar. This led to a paper published in a peer-reviewed scientific journal, co-authored with Dunbar.

Webber E, Dunbar RIM. 2020 The fractal structure of communities of practice: implications for business organisation

What it shows us is that the business world has a lot that it can learn from evolutionary psychology and natural social communities.


I have been working with communities of practice for several years. Back in February 2019, I set out to do some further research to see what insights would emerge.

What quickly emerged in the results is that the clusters of community size looked familiar. That is because they correlated with Robin Dunbar’s research on social groups and particularly Dunbar’s number. A theory that I often reference.

Distribution of Community of Practice (COP) sizes.
The X-axis is log10-transformed for illustrative convenience.

Robin Dunbar is a professor of evolutionary psychology at Oxford University. He has been researching the behavioural, cognitive and neuroendocrinological mechanisms that underpin social bonding in primates (in general) and humans (in particular) for many years. His story is fascinating, and I would recommend listening to this interview with him at to learn more.

Dunbar’s number is the cognitive limit to the number of people with whom one can maintain stable social relationships. It is often cited as ~150 people but breaks down into a series of numbers. These are 5, 15, 50, 150, 500, 1500, relating to close friends, good friends, casual friends, acquaintances and names you can put to faces. These groupings align with the amount and type of contact between people in the groups.

Although organisations are created, they have many similarities with naturally occurring communities. As social beings, we create social ties in the workplace, so it makes sense that communities of practice would have these similarities.

Because of this clear correlation, I joined forces with Robin Dunbar to get his take on the data and dig deeper into what it was telling us.

Five main takeaways

If you are interested, I would recommend reading the paper to get the full insights. Here are my top takeaways.

1. Business communities of practice mimic natural social communities

Communities of Practice seem to have a similar fractal distribution to natural social communities, as reflected in both the structure of personal social networks and the distribution of nested hunter-gatherer social groupings. This means that we can look to social communities to inform our workplace communities.

2. Communities of less than ~40 tend to have a greater sense of camaraderie

Smaller communities can create greater bonds and camaraderie. This has an impact on feelings of belonging and value derived from the community. Because of these bonds, they are more likely to meet regularly and be able to maintain the community without the need for a lot of organisational management intervention.

These feelings of camaraderie and therefore a support network can add to a positive experience in the workplace. Anecdotally, I have experienced this having a positive effect on recruitment and retention.

3. Communities of more than ~40 people need formally recognised leadership

Communities of ~40 people or less can be more democratic, with management responsibilities spread amongst members. Larger communities will tend to need formally recognised leadership. This number, which correlates with Dunbar’s number, is echoed in both natural human groups and constructed business organisations.

The question of leadership could be answered in many ways, including formal leadership roles, a core team and administrative support. All of which requires an organisation to formally recognise, support and reward time dedicated to community leadership.

The number of leaders in the community plotted against community size.
The X-axis is log10-transformed for illustrative convenience.

4. The larger the community the less often it is likely to meet

30% of the communities represented in the results meet weekly. The larger the community, the greater the gap between meetings. The mean intervals between meetings are 12.6 days for groups of 5, 23.9 days for groups of 15, 25.0 days for groups of 50, 46.3 days for groups of 150, 64.2 days for groups of 500 and 245.9 days for groups of 1500, which correlates to Dunbar’s ongoing research.

This has an impact on what a large community can do as a whole and points to the management overhead of bringing large groups together. It also has an impact on how knowledge and information can spread through the community.

Mean (±1 se) time between meetings as a function of cluster.
Note that, in a small number of cases, frequency of meetings was not specified.

5. Smaller communities or subsets in larger communities can carry out more focused activities and tasks

In free text, respondents in smaller communities stated that they took part in more varied and focused activities. It is challenging to have a face-to-face discussion or working session when communities get into the 100s or more. The implications for larger communities is to enable smaller groups to form within them for more focused activities and tasks.

You can read the full paper on PLOS ONE here

If you are interested in talking more about communities of practice or this research in particular, please get in touch.