Insight, Analytics and the New Marketing
Do influencers really exist?
Marc Earls’ blog had a link to this interesting video last week. Eric Sun from the department of Computer Science in Stanford talks about an analysis he did using data on 262,985 Facebook Pages and their associated fans. He analyzed how information propagates through the Facebook network by looking at the popularity of fan pages.
Sun wanted to test the hypothesis that the popularity of a page starts with a small number of important influencers who then influence their large networks. He also tried to predict how influential a person is based on the number of friends they have, their activity levels and their demographics.
The conclusions were interesting :
It is more important to find a large population on Facebook than it is to find a small number of influentials. The Facebook network is very connected. Because of this, ideas with good receptiveness will attract wide, long connected clusters.
Even if we wanted to find influentials, this would be very hard. Sun and his team did not manage to identify influentials based on activity levels, number of friends, Facebook age or demographics. Anyone can be an influential – it is only related to exposure.
If you are interested in the topic, the video is definitely worth 20 minutes of your time.
Also, the quants among you should check out other videos on the videolectures site. It’s a great resource if you want to brush up on your math skills.
[...] This post was Twitted by srikanththunga [...]
Hey guys, nice post – echoes similar findings by Duncan Watts’ (network theory scientist at Columbia University) whose findings with mathematical modelling also provide evidence against the Influentials theory.
Have a read of my post on it here:
http://infomagination.typepad.com/blog/2008/12/how-to-make-sense-of-a-complex-world-and-why-not-to-trust-malcolm-gladwell.html
Keep up the good work,
Matt.