Sunday, January 27, 2013

Are You Impacting the World or Just “Liking” it.

Your friends have more friends than you do.

It’s true. Don’t be hurt. Research it. Or rather don’t, and read on.

On average most people’s friends tend to have more friends than they do. This paradigm is known as the “Friendship Paradox,” and it is fairly recent to the social sciences. It isn’t a paradox at all, it is very real and there are mathematics to back it up. So what, who cares? We all should, and here is why.

In 2009, Harvard researcher Dr. Nicholas Christakis tested an experiment to research how the flu virus would spread through the university. He chose several hundred students at random and asked them to each nominate a friend. He monitored both of these groups separately and then compared them. What he found out, on the surface, was surprising. The friends who were nominated exhibited flu symptoms on average 16-47 days sooner than the students chosen at random. Because the nominated friends are more central to this social network, they were able get and in turn, pass on, the flu more rapidly. This is one of the most specific cases in the friendship paradox and its power. Here is a video of Christakis discussing his experiment.

Social networks are made up of people and their dyadic relationships. It isn’t the number of people nor the number of relationships that matter, but rather the interaction. The individuals on the periphery have the fewest direct connections within the network, while the people toward the middle exhibit greater centrality. Centrality is the notion of influence within the social network. The more centrality a person exhibits, the more relationships they share within in the network; therefore, the greater influence they have on diffusion within the network. Malcolm Gladwell expounded on this phenomena in his book, “Tipping Point,” and described these people as the social “mavens.” These are the people that really make the difference.
The social mavens exhibit the greatest influence on social capital. Christakis notes that if he were to have immunized the random group of students for the flu, very little, if any impact would have been made; conversely, had they immunized the friends who were nominated, they could have protected nearly the entire university at one-third the cost. This is astonishing stuff. While Christakis is mainly interested in using this phenomena to determine how to more easily predict epidemics, this same idea can be used to determine the influence of just about anything.

If one were to take a straw poll of her peer group, I could almost guarantee that all of them belong to at least 3 separate digital social networks: Facebook, Twitter, and at least one other (Instagram, Pinterest, Google+, etc.) But for all the “likes,” followers, comments, “plus ones,” what do we really have to show for it? What does it mean? Does the amount of followers, likes, plus ones, really determine how influential we are in our real lives, or have come to a point where our digital and physical lives are on ever diverging paths? We can watch the twitter-sphere explode with #whatever and Facebook posts blow up with likes and comments on the same topic. What if we could aggregate all these metrics into a measurement of social capital? What does 20 thousand twitter followers really equate to? Sure these metrics are easily quantifiable, but how do we qualify them as credible or as having clout? Speaking of c(k)lout...

Digital analytics are currently being used to summarize and determine who these people are and how they can be used to everyone’s advantage. is one such company who is attempting to determine a person’s digital social worth or influence within their realm of the social networks. Klout algorithmically aggregates all the different social networks you belong to. It consumes your followers, posts, comments. tweets, blogs, likes, pluses, hi-fives, shouts, whistles, and even Yahoo! accounts, then summarizes the data to determine your Klout score. The higher your score the more social influence you have in social media. On the Klout dashboard, it shows all your different social networks and what percentage of your score is determined from each social network, so you know on which networks you need to focus your efforts to increase your score. Here is a screenshot from my dashboard.

The scale goes from 1 - 100, with the average person scoring a 40. As an example, President Barack Obama scores a 99. Mine is a 12. Yup, i suck. Apparently I am only 10 percent as influential as the president of the free world. BFD. Omid Djalili scores 82 and he is a comic writer, producer, and self-proclaimed “dog handler.” I wonder how Ceasar Milan would score in that network? The basis of Klout is to determine your influence on driving social media and its trends. Does a tweet you make have a ripple effect-effect in the network, or did it fall on deaf ears?

While it is easy to quantify who is being “retweeted” and followed (#creepy), it is nearly impossible to determine if those people are actually influential in the same way the nominated friends were influential in the flu study. Sure my twitter feed blows up during sports games and elections, but i don’t consider the majority of these people influential. Being influential takes more than being a social network sounding board. To be influential you must have the ability to change the outcome of certain trends, norms or behaviors. While much of these theories remain constant across our physical and digital networks, one of the largest difference is the ability of the digital networks to actually pin-point, in real time who the mavens are. The influencers in a physical social network may have no idea how powerful their words and actions actually are; whereas, their digital clout can be viewed like scoreboard.

Gladwell, M. (2000). The tipping point, how little things can make a big difference. New York: Little, Brown and Company.


  1. Thanks Eric. I liked your post and am just curious to see if we can draw in any Italians and Germans to our Blog! I saw the stats are only one visit from Germany and None from Italy! Lets see if we can draw any attention from those two countries with a little German and Italian. I can only vouch for the Italian, but I hope the German is a decent translation.

    Venite a leggere un po' del nostro corso di Web Analytics!

    Kommen Sie und lesen ein wenig von unserer Web Analytics natürlich!

  2. Ich möchte gerne mehr Anmeldungen aus Deutschland für unseren Web Analytics Kurs sehen! Schauen wir was sich machen läßt! :0)

  3. Interesting topic. It is well written and kept me reading all the way through, so nice job. I signed up on Klout when I first heard about it about 6 months ago and could never really figure out how it actually worked. To me it just seemed like the way you got more points was by either adding a social network or getting friends to join the site, which seemed kind of pointless. It is a good idea though to track your online social influence because some people do actually influence change but most just make a lot of noise.

  4. Apparently I'm not the only one who found the Klout scoring system to be interesting & helpful. I mentioned it in my blog too. Great website & ideas. I agree with what you mentioned regarding the difference between people that are truely influential and others that just have a large a online following or a lot of online friends. I can see how some people can appear very influential & important yet end up being a total waste of time for companies to target.