Klout Adds Facebook Data to Its Influence Graph
As millions more people tweet, blog, comment and update their status every day, the difficulty of finding signal in all that noise increases exponentially. Some services try to filter all of that social-media content by keyword, some use location, while still others use algorithms to determine whether someone shares your interests. Klout, a two-year-old startup based in San Francisco, is trying to build a database of influence that can determine how influential an individual user is on social networks — that is, how much impact the content they create has on others around them. The company today announced that it has expanded beyond its traditional reliance on data from Twitter and is now indexing content from Facebook as well, with a view toward becoming the dominant measure of influence “across the social web.”
Measuring influence isn’t just something that Klout wants to do in order to make users feel good about themselves, or so it can give them badges for passing certain milestones (although it does that as well). The reality is that as social media and social networks have become a larger and larger phenomenon, marketing agencies and companies have become increasingly interested in using these networks and services to target specific demographics, and to target “influencers” within specific topic areas who can help spread their message. So the Palms Hotel and Casino in Las Vegas is using Klout scores to build a group of influencers it can approach with special offers, and Virgin Airlines offered special rates to Twitter users with high Klout scores.
Twitter is obviously only part of the social web, however, and until now, Klout has been restricted to measuring influence solely on the basis of tweets and re-tweets. Data from potentially hundreds of millions of Facebook users could change that dramatically (provided they want to connect their Facebook accounts to the service, that is). In a blog post, Ash Rust — the company’s “director of ranking” and a former “relevance engineer” at OneRiot — says Klout has been working for several months to analyze Facebook accounts and determine how influence functions on the social network, which sees more than 30 billion pieces of content shared every month.
Unlike some services that simply look at the number of Twitter followers or Facebook friends a person has, Rust says Klout looks at a user’s “ability to drive action,” or their ability to influence others to click and say they “like” something or share a piece of content, post a comment, etc. On Twitter, the service looks at statistics such as the number of followers a user has, or how often their messages are re-tweeted, but also looks at several different levels of influence — for example, how often a user’s messages are re-tweeted by other users who have a high level of influence. The company then ranks a user on metrics such as “reach” and “amplification,” as well as the size and influence of their larger social graph.
Rust admits in his post that Facebook allows users to “interact with their friends in more complex ways than we’ve previously seen,” and that it will take some work to come up with an accurate picture of what that means. The payoff for coming up with an authoritative ranking of influence across two of the largest and fastest-growing social networks could be substantial, and Klout is already partway down that road, since its data is already integrated into a number of leading Twitter tools and services such as HootSuite and Seesmic. Other companies in the same space include PeerIndex and TwitterGrader.