Dunbar's Number and Project Management

In the late eighties Prof. R. I. M. Dunbar, in his anthropological studies, discussed optimal group sizes for primates, which seemed to be a function of relative neocortical volume in their brains. According to Dunbar, the number of neocortical neurons limits the organism's information-processing capacity and that this then limits the number of relationships that an individual can monitor simultaneously. When a group's size exceeds this limit, it becomes unstable and begins to fragment. This then places an upper limit on the size of groups which any given species can maintain as cohesive social units through time. In other words, there is a natural limit beyond which the group's functionality and efficiency begins to deteriorate. For humans, the Dunbar's number seems to be around 150 (Dunbar actually calculated the mean group size to be 148). Dunbar also noted that the group has to be highly motivated for such a large number to work - in the absence of intense economic or environmental incentives and "social grooming" such large groups tend to break down into smaller fragments.

Here are a few thoughts on how his study impacts the way work groups should be structured.

• The Dunbar's number indicates the total number of relationships a person can maintain, including family, friends and professional relationships. Many authors on the web take the number to mean the optimum number of people that you may have on a project. I think there are two mistakes these authors are making here. (1) They are confusing the maximum with the optimum. The brain is wired to monitor up to approximately 150 persons. This is not the same as saying that the brain monitors 150 people best. And, (2) that the 150 relationships are exclusively for the work environment. If you take into account relationship slots that may be occupied by friends, family members and professional relationships outside the project environs then an appreciably less number of slots are available for project related relationships. A Facebook study from a couple of years ago estimated that an average user tracked about 40 "friends". Add to this about 20 to 25 people that a person may relate to in some other way but who are not members of Facebook and this will leave a maximum of about 80 to 90 people that a team member may effectively monitor. The optimum number will probably be some where about half of this.

• A macro group may be formed which may be composed of many smaller cohesive groups, each containing the optimum number of members. These smaller groups may communicate with each other through the upper hierarchy levels, which forms the communications/coordination backbone of the organization.

• The larger the group the more "social grooming" that is required. In other words, larger groups need more team building exercises to keep them functioning as a productive whole. According to Dunbar, for numbers approaching the limits the time required for these activities may be as much as 42% of the total time. This means that there is a diminishing rate of return associated with increase in team size. This is another reason why the actual optimum number of team members in a group will be appreciably less than Dunbar's number.

Interesting read: http://www.wikinomics.com/blog/index.php/2009/06/15/diminishing-returns-of-collaboration