I’ve said it before and I’ll say it again: influence is the social media industry’s white whale. Every data scientist and pundit has had a crack at explaining it.
You don’t have to trawl through our explanation of influence if you don’t want to. By all means, jump through to our rather beautiful, interactive network visualisation of the top 200 most influential companies and individuals within New York City’s social media community. Click it. Drag it. Marvel at the nodes and edges and colours. Search for yourself (in the map, not existentially).
Mostly, we just wanted to get to know you all. We’re moving to New York pretty soon, and we wanted to say ‘hello’.
If, however, you’re as much of a data geek as myself, here’s a brief overview of the science behind the glamour (or should we say, glamor).
To make our dashing dataviz, we crawled all the relationships between users on Twitter who identified themselves as being based in New York. Armed with such an extensive social network, we were able to run all sorts of fancy network science to discern some rather interesting things.
Firstly, influence, which is represented by the size of each ‘node’. In this case, we define ‘influence’ by the number of incoming connections a person or organisation has from other ‘nodes’ within the NYC social media community. And not only the number of relationships, but also the influence of the ‘nodes’ on the other end of those relationships – so the more influential people that pay attention to a given person, the more heavily we weight influence for that ‘node’. We think this form of ‘peer appreciation’ is a good proxy for the relevance and importance of a community member.
Secondly, we can discover sub-communities, identified by a node’s colour. A metric called modularity automatically finds tightly interconnected clusters that form distinct groups within a community. In our analysis, we found the following types of groups:
PINK: Agencies, services, events, NYC mayor and his office, PR , online magazines
BLUE: CEOs, founders, VPs, Directors
GREEN: Professionals in the digital and online social space
ORANGE: Bloggers, authors, journalists and editors
Of course, they aren’t exact as they are automatically coloured based on the connections a person has to others. In many cases a person may be strongly connected in more than one community and hence the colour may reflect the extent to which a CEO has links to journalists and editors, for example. But it’s pretty good for an automatic segmentation, right?
Now, this map is just a bit of fun, and we hope you enjoy it. However, mapping social networks can be particularly helpful for brands looking to identify communities their fans/potential fans belong to, as well as the influencers within those groups. Our in-house tool Mappr (which I coded powered by far too much caffeine) makes this unique approach quick and easy. If you’ve got questions about the methodology, the data visualisation or just want to chat social insights, drop me a tweet at @mickyconroy.
Now go on; take a look around our map of the 200 most influential companies and individuals within New York City’s social media community.