Five Truths about Having a Networked Life

I recently completed “Networked Life” – a six-week course taught by University of Pennsylvania Professor Michael Kearns, the Founding Director of Penn’s program in Market and Social Systems Engineering.  This popular course, with more than 40,000 students through Coursera.org, looks at “how our world is connected socially, strategically and technologically – and why it matters.” It draws out the common and predictable behaviors of social networks as well as the behavior and performance of the individual members. These are my top five of the take-home points that may have practical applications in business:

1)      Network structure impacts outcomes. Knowing the structure of a network makes it possible to allocate resources more efficiently. This is true, for example, in preventing the spread of disease or marketing a new product. Of course, it is very difficult to know the precise network structure of most real life networks. Network structure also plays an important role in the distribution of wealth and resources. A “clan style” network, where there are few connections between clusters of members, results in the greatest disparity in the distribution of resources.

2)      The theory of six degrees of separation holds true.  Just like with any actor and Kevin Bacon, the degree of separation between any two individual members of a large network is typically no more than six people.

3)      More connections can mean more social media value.  Some small number of members of a network typically have a much higher number of connections than average. This function is typically displayed on a graph as a bell curve with a heavy right tail. From a marketing perspective, for example, these few members in the “tail” may be viewed as having significantly more social media value.

4)      Variety in connection locations helps.  A mixture of local and long distance connections are required in order for the members of a network to navigate efficiently across the network, make connections, and engage in transactions most effectively. These network properties, often desired, are termed “high clustering coefficient” and “small diameter” by researchers.

5)      Science +  Math + Networks can predict behavior.  Overall, the members of a network perform very well in achieving an even distribution of wealth and resources. But in some cases even when the mathematical models predict an even distribution of wealth and resources, real life experiments with humans result in a high variance in wealth and resources. The larger the wealth or resource variation predicted by a mathematical model, the lower the overall behavioral performance of the human members of the network.

About the Author

Tony Novak, CPA, MBA, MT is a sole practitioner based in Philadelphia area with a background in online publishing. He was the founder and former principal of several Web-based financial services including MedSave, Freedom Benefits, OnlineAdviser and OnlineNavigator. Now he focuses on financial planning services for attorneys and can be reached at tony@tonynovak.com .