Three weeks ago, our class touched upon the case of Todd Williams and social networks and brokers. The Todd Williams case dealt with the eponymous financial analyst who recently joined a Biotech company is caught between the conflicting requests of two managers in an unfamiliar organization and feels helpless. As we discussed in class, the case is a good parallel of corporate organizations and I wondered if the predicament of Todd was due to the lack of a ‘network’ of mentors within his organization and thus a lack of power. The topic of networks science and its effects in organizations and in life is fascinating due to the prevalence of data. Over the last decade, there has been an explosion of user data from collected from social networks – Facebook, LinkedIn, YouTube, Google, and Twitter, and new methods to analyze data. I wondered what are the requirements for social networks to be useful and are larger networks necessarily better?
In my research on the topic, I came upon an article written by two physicians on the Harvard Business Review arguing a larger network is useful because it provides data for mathematical analysis of people’s role (‘You can see who’s connected to whom. You can determine who’s a hub and who’s a mere outlier’) and thus allow salespeople to target specific doctors who are the most connected. Studies have shown that in the introduction of the drug Januvia, physicians were more likely to prescribe the diabetes medication if they had Januvia adopters in their networks which extended to three degrees of separation (a colleagues’ colleagues’ colleagues). Finally, a physician who has a larger social network possess a further advantage. The authors describe the sale of Lipitor, a cholesterol control drug, declined when a generic was on the market illustrating that ‘interconnected doctors switched their prescriptions almost simultaneously, like a flock of birds changing direction’ and was able to help patients save money. Thus, a larger social network does seem to be better for physicians as it allows faster introduction to new and more effective drugs, as well as and cheaper generic drugs.
Furthermore, I found an article published by Michael Simmons, a business columnist for Forbes Magazine, where he described the impact of network science for professionals. What Mr. Simmons found was radically different and he says that being the most connected is merely a ‘vanity metric’ and cited research from Rob Cross, a leading applied data scientists from the University of Virginia. A video describing Cross’ research is shown below:
One of the most important distinctions that Cross and Simmons makes that is information is a set of clusters, not a larger cluster. The implication of this is:
- Information travels quickly within a group: the same information is quickly spread and repeated in a group. As a result a group develops a ‘shorthand’ for information
- Information does not travel between groups: information becomes ‘sticky’ and becomes harder for other groups to understand the full meaning because of the ‘shorthand’ used by one group
Old Paradigm vs. New Paradigm of Social Networks
The implication of this model is that a social broker is a translator and is able to convey the full meaning of information between two groups. Simmons further cites Ronald Burt, who is considered the father of network science who says: ‘What a broker does is make a sticky information market more fluid. Great ideas will never move if we wait for them to be spoken in the same language.’
In a novel research by Ronald Burt, candidates with an ‘open network’ compared to a ‘closed network’ have a greater career success (all else being equal). Burt attributes this to the fact that in an ‘open network’, people do not know each other yet and a broker can be between many different clusters. In a closed network, everyone already knows each other. As you can see in the chart below, the further right a person tilts towards closed network, the more a person will receive the same information which limits the utility of information. The further left a person tilts towards open network, the person can receive new information and thus become brokers. People who belong to ‘open networks’ on the left are significantly more successful than those to the right.
Relative Performance of Individuals vs. Network Constraint
Thus, for social networks to be useful there needs to be a good structure as well as a large network i.e. it’s not about how big it is, it’s about how you use it. A large network is useful in order to generate the necessary data and to find the person who can become the translator. However, success in a career is strongly predicted by the type of network and ‘open network’ allow individuals to receive a greater amount of information necessary to become a network broker. Thus, in Todd William’s case, the organization he worked for had a closed structure where information was kept closed at the detriment of the overall performance of the organization. Finally, a good network broker is not only a person that can relay information between two camps but also is a capable translator who can decode the group speak of each group and in turn allow the two groups to collaborate.