If you don’t know www.klout.com you are probably not really into social media. Or advertising. Or influence measurement. It is meant to measure how much you influence people. And about a year ago they changed their algorithm. It was meant to get better. But it didn’t.
But how do I know that it got worse?
Well, for starters I noticed that my Klout score changed. Since I didn’t change anything dramatic on my personal social media accounts it was obviously them. A little fishing around showed that Facebook had been seriously upgraded in terms of weighting and Twitter downplayed. This was counter-intuitive. It still seems silly. Surely a more public domain like Twitter where people aren’t obliged to “like” something simply because you know them is more objective a measure? And retweeting is generally a much weightier backing of a though than Facebook sharing. There is much less pressure to stay attached to an account on Twitter, you can unfollow fairly easily.
So how can I measure how “wrong” I think Klout is? (Let alone that Facebook paid them to rig their metrics.) Remember, this is a measurement that many other businesses rely on. Many media monitoring tools have it on the x axis of “social media influencers”. You want to know who is talking about your brand or business and matters? Eh, well, it might be based on something which doesn’t work too well any more.
There were two ways for me to look into this. One was using different accounts. I have access to a lot of Facebook and Twitter profiles. I experimented with older accounts (many with 5000 “friends” or 2-3000 followers). With Klout you can associate your Klout to any combination you want. So I would start off with one Facebook account. Let it settle down….Klout of 55. Associate a Twitter account….Klout goes up to 56. Ahem….. Unlink the Facebook account. Measure again. Link a different Facebook account. Klout doesn’t seem to mind because obviously it wasn’t designed for data maniacs like me trying to reverse engineer it.
The other way was to get people I know well to join Klout. People whose Facebook or Twitter habits I understand in depth. And this is were the current Klout algorithm lost any respect I had for it. It is fairly easy to increase the level of interaction a real user with a real Facebook profile has with his Facebook friends. Facebook hasn’t addressed this issue so much because they are too busy focusing on Pages and Promoted Posts and all that. So friends who are active on Facebook, especially when they are photo heavy in their posts, can get ridiculously high Klout scores right from the start.
There is another reason I assume that Klout have got it wrong: people are too scared to tell them! “What if they see this post and downgrade my score?” Oh no! Nobody will take me seriously anymore!….
Influence measurement is serious business. If a fake Facebook account which simply reposts stuff, or a friend who is just a decent photographer can easily hit a Klout of 65, it is probably time for somebody to take it a bit more seriously. Or just come out and announce it officially that Facebook is funding Klout.