Influence on Twitter is not necessarily tied to the number of followers one has, at least this is one of the conclusions in this report on Twitter influence which came to my notice by a @papadimitriou recent tweet. The core finding: it’s more influential to have an engaged audience that actively retweets and mentions the user, and one keeps this audience engaged by frequently interacting with this audience (see common sense paragraph below).
The dataset used in the report consisted of nearly 2 billion follow links among over 54 million users who produced over 1.7 billion tweets. Here’s a link to a website the report authors set up that further describes the dataset. The report compares three measures of influence: indegree (i.e., number of followers one has), retweets, and mentions. The most followed users were news sites, sports stars, and politicians. The most retweeted were content aggregation sites like @Mashable and @TwitterTips, @GuyKawasaki, and @TheOnion. The most mentioned individuals were mostly celebrities.
The report found that retweets are primarily content driven (including both the username and link to a source), whereas mentions are mostly identity driven. In looking at retweets and mentions, individuals who were both retweeted and mentioned the most showed the highest engagement metrics (i.e., conversing with their audience), as well as posting creative and interesting tweets.
Finally, the report focused on how much influence “ordinary users” can exert within the Twittersphere. In this case, the authors looked at how completely unknown users exerted tremendous influence (as measured by retweets and/or mentions) during serious news events. By limiting their tweets to a single news topic, users like @oxfordgirl quickly gained influence.
This report, I suppose, just reinforces common sense at some level. But it’s nice to see someone crunching the numbers to reinforce one’s intuition.