Everyone’s talking about “social”: “social networks”, “social media”, “social web”, “social games”, “social apps”, social everything, increasingly culminating in the idea of “social business”. But what does it mean to be social in this context? Are you more social when you have 1,000 followers on Twitter or Facebook rather than just 100? It certainly seems that way when looking at many social media success stories. However, can you actually be social with 1,000 or more people? Robin Dunbar, Professor of Evolutionary Anthropology at the University of Oxford, believes that the cognitive limit to the number of individuals humans can maintain social relationships with is at 150. Much to the professor’s surprise, “Dunbar’s number” has made a considerable career on the Internet, certainly helped by the popularization of his findings through Malcolm Gladwell’s book “Tipping Point” and supporters such as marketing guru Seth Godin. Still, Dunbar’s number is not yet common knowledge, and among those who know it, it’s meaning is controversial. So, I figured it’s worth looking into the concept and its possible applications for business communications.
Based on studies with primates showing that there is a correlation between the size of their social groups and the size of their brain’s neocortex, Dunbar speculated in a 1992 paper that this correlation could be extrapolated to humans and predicted a group size of approx. 150. He then compared this number with data on social group sizes in human history such as the sizes of villages, tribes or units in ancient armies and found his hypothesis confirmed (within a band of variation). In a 2002 study, he also examined the “social network size in contemporary Western society based on the exchange of Christmas cards”, and again, the “maximum network size averaged 153.5 individuals, with a mean network size of 124.9”.
If 150 is indeed the maximum of stable inter-personal relationships humans can maintain, Dunbar’s number also explains why larger groups don’t work without additional structures such as hierarchies or laws to organize the relationships. This is what Malcolm Gladwell picked up on pointing out that organizations suffer from a sharp productivity loss when growing larger than 150 employees. As a result, companies such as Gore, the manufacturer of Gore-Tex fabrics, organized their teams in smaller groups, so that they could keep strong working relationships.
Social GroomingFinally, managing more than 150 stable relationships would not only go beyond the brain’s capacity but also take too much time. In that regard, humans face the same challenge as primates: social grooming is laborious! It has been suggested, though, that humans might be able to increase the number of relationships with the help of social media, since it makes it easier to memorize a relation’s origin and then stay in touch, even over large distances. Dunbar counters, though, that dependable relationships need in-person meetings and social media still doesn’t remove the biological constraints (see for instance this video interview Dunbar had with the Guardian). And indeed, a recent study by Gonçalves, Perra and Vespignani (August 2011) validated Dunbar’s argument. They modeled activity of 1.7 m users on Twitter over six months and found “the data in agreement with Dunbar’s result; users can entertain a maximum of 100-200 stable relationships.”
Does this mean any number of social network connections larger than 150 doesn’t make sense? Well, it depends on what you are trying to achieve. For instance, a mobile social network company called Path limited the number of friends users can have on their network to 150 (with reference to Dunbar), since their service is built for circles of close friends. That’s understandable. However, as early as 1973, sociologist Mark Granovetter pointed out that not only those people you are closely connected with are important for your network. In fact, they might even bring less value to it, because they probably know the same people and things that you do whereas your wider network of weak ties might open up access to new information and new networks. On a larger scale, the component of weak ties supports the concept of crowdsourcing as famously crafted by James Surowiecki back in 2004: If you really want to extract the wisdom of crowds, diversity is one of the critical success factors, and you will only get it out of a wide network with weak ties.
So, from my perspective, Dunbar’s number is certainly a factor business communicators have to take into consideration. If you want to have a group of people interact and collaborate productively to get things done, the size of this group shouldn’t go beyond Dunbar’s number. Making this group larger might lead to a loss in productivity, because an increasing amount of “noise” in the system can be distracting and will increase the need for coordination and filtering. Also, large numbers of followers or “fans” shouldn’t be mistaken as a proof of social interaction that is by default allowing for stronger relationships than traditional reach. Dunbar has shown that stable interpersonal relationships can’t be managed in large numbers, unless you do it one Dunbar group at a time. IBM employees, for example, operate thousands of internal and external social media platforms creating tens of thousands strong ties inside the organization and outside-in.
Conversely, if the goal is to open up a specific business unit or the whole organization to innovation, it is necessary to go far beyond the Dunbar number and include weak ties. In fact, too many strong relationships within the crowd you are working with will threaten the diversity needed to go beyond existing approaches. Online communities in particular are threatened by the “echo chamber effect“, because once a claim is made by one participant it’s extremely easy to have it repeated by like-minded people, resulting into the reinforcement of their beliefs and possibly hindering critical discourse. Memes or myths are being created as a result and can block the way for innovation. In addition, the echo chamber effect can be aggravated by algorithms homogenizing search results within social networks, a phenomenon Eli Pariser called the filter bubble. In fact, I sometimes do have the impression that we as business communicators do suffer from the echo chamber in our social networks, because a Dunbar group is reinforcing myths such as the loss of control on the social web at conferences and on the web. Thanks to Dunbar and others, we can better understand these mechanisms. It is a whole different discussion how to create a wide network with weak ties allowing for diversity and then aggregate the ideas for innovation. Perhaps, a discussion for another post. This one has already been long enough!
Georg Kolb