Yesterday, we had some fun looking at Fast Company’s Influence Project. (Well, I had some fun. If you did too, well, that must be my influence at work, right?)

Influence is weighing heavily on the social media community’s collective mind right now. I’ve noticed a surge in Twitter and blog chatter around Klout, a tool that aims to measure influence on Twitter.

Klout’s a pretty sophisticated tool, going way beyond a follower count:

 

Klout uses over 25 variables to measure True Reach, Amplification Probability, and Network Score. The size of the sphere is calculated by measuring True Reach (engaged followers and friends vs. spam bots, dead accounts, etc.). Amplification Probability is the likelihood that messages will generate retweets or spark a conversation. If the user’s engaged followers are highly influential, they’ll have a high Network Score.

We believe that influence is the ability to drive people to action — “action” might be defined as a reply, a retweet or clicking on a link.

The reports are detailed and interesting… and I’ll admit I’m also charmed by the way Klout elaborates on its assessment, using phrases that wouldn’t be out of place in a job application letter, performance review or LinkedIn recommendation: “@RobCottingham has worked very hard to successfuly build a large, highly engaged network.” “@RobCottingham is effectively using social media to influence their network across a variety of topics.” (Hey, thanks! It almost salves the wound of having dropped nine points in the last week.)

Klout’s newfound, well, influence has landed it a special place in the heart of Twitter web app HootSuite. (By the way, I’ve been using HootSuite heavily for the past few days, and can report it’s amazing.) You can now filter the people you follow so you only see updates from those who surpass a certain Klout threshold.

 

That’s a great way of seeing what the A-listers you’re following are talking about. But if you’re going to use that feature, may I recommend a healthy dose of caution?

There’s a risk with measurements like these that they become self-fulfilling prophecies, and reinforce attention hierarchies. If enough people use Klout to divide the world between the influential and the non-influential, and listen mainly to the former, then the influential will continue to be influential – because audiences aren’t hearing other voices.

And you’ll miss out on some below-the-radar surprises. Because as cool as Klout is, it doesn’t take into account the fact that your influence on Twitter is going with one segment than with others. It doesn’t account for long-tail phenomena: people who are leaders in their particular communities. And it doesn’t account for the kind of influence that isn’t so easily measured automatically.

Ultimately, Klout gives you one number – derived from many factors, true, but it’s a single number, aiming to measure something that is insanely multidimensional. I don’t want to take anything away from what they’ve built – it’s a great tool, it’s elegant, it’s beautiful, it’s engaging, and I can see myself obsessing over it in the weeks to come – I really want to get more badges – but don’t let it dictate where you direct your attention.