Identifying New Influencers

As time goes by there is more research published on social media and social networking use. Here are some examples on identifying and measuring influence.

Research firm Sysomos has released some interesting data on Twitter use. In round numbers,10% of Twitter users make up 90% of the activity and 1% of Twitter users update more than 10 times per day. Approximately 90% of Twitter users post less than one update per day. Only 1% of Twitter users follow more than 1,000 people. About 90% of twitter users follow less than 100 people. Now this is all consistent with Dr Jacob Nielsen’s 90.9.1 rule:  “In most online communities, 90% of users are lurkers who never contribute, 9% of users contribute a littel, and 1% of users account for almost all the action”. As I’ve mentioned before, the 90.9.1 rule is a well-established pattern in social groups.

There are a couple of observations here. One is that the term “hyperconnectivity” needs to be taken in the context of how much attention is paid to “followers” or “friends” on social networks. Facebook’s own analysis shows that, while the average Facebook user has 120 friends, people maintain regular contact with only about five or six those. The rest are “weak ties” as it were. Those people who are highly active tend to have more connections (eg.1% of Twitter users are most active & 1% follow more than 1,000 people). The two may not be the same set, but based on my own use of Twitter I would say that they are the same. Sysomos data backs this up too: once someone has 1,000 followers the number of tweets/day rises from three to six.

Here is another piece about Twitter, this time as a leading source of real-time information. Time and again Twitter has demonstrated its strength as a source or real-time eyewitness news, heading off  the big media brands, this time in Iran. But there is a downside. To quote Inquisitr “When recognizing Twitter as the best outlet for the latest on the ground, you can’t ignore that the data coming from Twitter is raw, unfiltered, and at times difficult to follow”.  This is where adding value comes in through channelling attention to particular content. According to the Inquisitr article, that is exactly what the Huffington Post (an online newspaper) did to good effect in using data on post-election Iran that had been distributed over Twitter.

Nielsen has reported that time spent on Facebook in the US increased from 1.7 billion minutes in April 2008 to 13.9 billion minutes by April 2009, a growth rate of 700%.Time spent on the top 10 social network and social media sites increased by 212% over the year to April 2009. Interestingly, time spent on MySpace was down 31%. Some previously popular sites (such as Friendster) are no longer in the top 10. The assumption is that users find it easy to migrate from one site to another. I’m not so sure that it’s as simple as that. Facebook now has the largest number of photos of any site and there is no easy way for people to take their data with them right now.

MySpace seems to be more popular with teens, and more popular for viewing online video. Nielsen reported that: “In April 2009, visitors aged 25 to 34 and 35 to 49 were the highest indexing age groups on Facebook, being 27% and 23% more likely to visit the site than the average user, respectively. In contrast, the highest indexing demographics on Myspace.com was people aged 18 to 24 and 12 to 17″.

Identifying social media influence looks like it will be no trouble: focus on the top 1% of users (by activity and number of friends or followers).

The hard part is in measuring influence. According to the Sysomos sample, ‘Twitterfluence’ (as I call it) adds up to about 120,000 people. Time spent on sites is another measure of influence, a more straightforward task. Demographic analysis is another way of measuring influence. For example, tracking teenager social activity would involve more analysis of MySpace relative to Facebook. Tracking the use of social networks like Twitter by other media brands and by individual bloggers of note would also be another measure of influence.