Edelman Takes on Klout in Social-Influence Game
Is it premature to say that Klout has become the de facto standard for measuring a person's online influence? Probably. But the 3-year-old San Francisco-based startup sure has defined the terms of the discussion. Right now, there are debates whether the Klout score should be added to resumes, some of you are eyeballing your count the way a diabetic watches his blood sugar, and this fellow even went to the trouble of issuing a press communique merely to announce a fondness for the whole enterprise.
Why ascribe so much power to a service that isn't much more than a tabulation of how many followers and retweets you have? The all-too-human explanation goes like this: People like scores, which Klout gives them. People also like gifts, which Klout gives them if they're good enough at playing Twitter and Facebook and LinkedIn. Marketers such as Audi and Starbucks are taking part in its Perks program, and that gives Klout revenue. Everyone's winning -- everyone except people whose vision of influence doesn't cotton to gamification, swagification or oversimplified math.
Among those are the PR colossus Edelman, which actually got into the business of monitoring social influence in 2009 with much less fanfare when it launched its Tweetlevel tool, Tweetlevel.com. Tweetlevel went down earlier this year thanks to an API change from Twitter that affected a lot of third-party apps, so Edelman took the opportunity to revise it as well as Bloglevel, a complementary service. Its London-based creator, Jonny Bentwood, promised me, "This is not a game. There are no perks. And there is no way to game the system."
I spent the past couple of days trying them out and was impressed by the direction the firm is taking, if not as much by the current experience of using the services. These are tools that manage to be serious and useful to a marketing person who wants an actionable understanding of influence around specific topics without being so complicated as to require several days of training prior to use. That, however, does not mean you should rush to use them. The sites, which are just today coming out of beta, are very buggy and very slow.
What's probably more important than how they can be used right now is where they're trying to take social influence. The key differentiator for Edelman's tools is a focus on topicality. While Klout will show any user the topics he or she on which he or she is influential, it doesn't facilitate searching on general topics of interest or on fast-moving news stories.
So your Klout score is relative to everyone else on Klout, and we're all competing with Justin Bieber and his perfect score of 100, regardless of whether we're teenager years old or have a fevers named after us. There's little to no context, just a formulation of Twitter-derived popularity. And as Mr. Bentwood told me, "Influence without context is meaningless."
While Tweetlevel does include vanity trappings like overall influence rankings and badges, it's how-to argues you should begin with a topic search. So I tried to see who's saying what about the News of the World saga. After a short but noticeable wait, I got info on the volume of chatter, popular terms being used, and a rundown of the most relevant users. Then I got a ranking of the most influential media brands or personalities relevant to the topic. In the top 20 were Twitter all-stars like Neil Gaiman and Stephen Fry, media thinkers like Jeff Jarvis and GigaOm's Mathew Ingram, and a host of lesser-knowns -- a pleasing mix of the expected and the surprising. A search on "Dear Netflix" yesterday showed a much more grassroots-y field of influencers helping to spread the rage at the company's price hikes, with most of the relevant influentials unknown, at least to me. I could see this being useful to a PR person in the midst of one of these crises.
As I mentioned, there were problems. Trying to search across both Bloglevel and Tweetlevel at once was wobbly, to say the least. Another problem is the unwieldy process of getting from a relevance ranking to an influence ranking that involves receiving an email. Mr. Bentwood changed this for me on the fly, but it was unclear how this will look in the future. This was one of many such clunky interactions that weighed things down.
When I complained, Mr. Bentwood didn't fight me. "I think you are right," he replied. "The [user experience/user interface] can be better, which is why Edelman is committed to the long-term development of our Level tools." He went on to explain that this will involve on-the-go tweaks and quarterly upgrades.
Edelman's roadmap includes Levels for Facebook, YouTube, LinkedIn, offline and a way of aggregating them all that , in my eyes at least, is sort of the holy grail when it comes to measuring social influence. All this platform-specific stuff is excruciatingly artificial. Looking only at one rather easily measured platform as a proxy for online influence such as Twitter ignores the power of others, whether a social network or a traditional media brand. In these sorts of rankings, David Carr certainly gets credit for his enormous Twitter following, but can the same be said for his central role at The New York Times?
I have my doubts as to whether this problem can be solved by wholly automated systems. Something so complex as influence screams for a touch of human judgment. For those whose faith does lay with the algorithm, it's good news that the likes of Edelman is trying to smarten things up. Now if only it could borrow of some of Klout's site experience.