Remember the great website engagement debate?

Remember the great website engagement debate? Remember all the hand-wringing and byte-spilling about formulas that were supposed to represent the holy grail of web analytics? Mercifully, the echoes of those debates have subsided. But now something new, and equally inane, has emerged: the concept of twitter influence.

Not content to merely enjoy the messiness of this wildly beautiful medium, adherents of the twitter influence movement posit that you can derive an accurate reading of tweeter’s influence by looking at a series of metrics associated with his/her twitter account. The formulas vary, but they all contain some witch’s brew of the following metrics: follow, unsolicited mentions, re-tweets, presence on lists. Some also contain fractional calculations that measure the relationship between two of those metrics; the lists to followers ratio is one I’ve been reading about lately.

Here’s the problem with all this: in a medium where the barrier to participation is literally zero, how much stock can you put in any behavioral metric? A celebrity merely has to list themselves and be verified, and wham! they’re up to 50,000 followers overnight (remember how quickly Oprah got up to speed?). Brands get followed copiously, but only 23% of their followers do it because the tweets are interesting or original . Re-tweets are more about link circulation than actual influence (57% of retweets contain a link versus 19% of normal tweets). List reach is a derivative of follower reach (Ashton has been listed over 12,000 times already). And a high ratio of followers to following is more of a barometer of snobbery than influence.

While impressive metrics might land you a TV deal based on tweeting what your dad says, it doesn’t earn you influence. I challenge the disciples of twitter influence to tackle this from a different angle. Ask people whose tweets they wouldn’t be able to live without? On whose recommendation would they be most likely to rush out and buy a book? Would they switch their browser or OS on the basis of someone’s tweet?

I challenge them to tie it back to a real world event. If they can prove to me that Guy Kawasaki’s tweets move people to take real, concrete actions, then I’ll grant that he has influence. Anything less than that, and they’re just measuring engagement.

Jonathan Levitt