The social media bits of the Twittersphere I read have been very approving of an AdAge piece this week by Matthew Creamer. Creamer lays into the firm Klout, who peddle a metric of how influential someone is based on the Twitter activity surrounding them. In a press release issued at the end of the year, Klout announced that the most influential person on the web, the sole individual to receive their perfect score of 100 in influence terms, was…. Justin Bieber.

Cue a predictable storm. Bieber was more influential than Obama, than the Dalai Lama, than – well, anybody, obviously. Creamer’s piece is smart about Klout in general, about Twitter as a self-contained game system, about the foolishness of using Twitter followers as a baseline metric for anything, and about a lot of other things. But I found what he had to say about Bieber interesting. Bieber, he explains, isn’t influential, he’s just popular. Influence is the ability to drive action, and what can Justin B actually do?

It’s hard to imagine that Justin Bieber, with his 6.4 million followers, is driving much behavior other than getting people to talk about Justin Bieber, frenetically retweet him, and possibly buy a record. Is that influence?

Erm… hold on? This is the Justin Bieber who is a pop star, yes? So what action is he meant to be driving, exactly, beyond being talked about and selling records? He’s sold over two million of them in the US this year – you try doing that these days! If they’re all on Twitter that’s a roughly 1 in 3 conversion rate which sounds well worth a high score on anyone’s metric.

I’m not especially a Bieber fan, I’ve never heard his album. I’m just using this as an example of the main problem with influence: not that people are measuring it wrong, but that the word is so apallingly fuzzy that most of the problem in measuring it comes from a complete failure to define what it might mean. (This line of thinking won’t be novel to anyone who’s talked to FT’s Mark Sinker on the topic). Influence is not popularity, says Creamer. Quite right. But if it’s the “ability to get people to do stuff” then as we see Bieber has plenty of it and Obama – whose social media goal for 2010 was presumably “get people to vote for my party in November” – doesn’t have quite so much.

The problem with the “get people to do stuff” idea of influence is that people have very different spheres of influence, so inevitably you end up having to step outside your metric to make a judgement about what kind “do stuff” matters more. The piece posits marketing thought leader Seth Godin as an example of someone very influential who doesn’t use social media. It’s true that Godin’s ability to get other marketers to repeat his ideas seems on a similar scale as Bieber’s ability to get fans to act on his tweets: but if Bieber is just “popular”, why isn’t Godin?

Another expert quoted talks about influence in terms of authority and trust. Yet another talks about it in terms of “thought-provoking ideas”. Now, none of these are stupid or wrong things to be interested in, the problem comes if we start using the same word for all of them. If Klout is talking about popularity, and Creamer is talking about power, and someone else is talking about authority, and someone else is talking about thought leadership, then why are we using the word “influence” at all instead of the words “popularity”, “power”, “authority”, etc.?

I’ve found probably the most useful way of thinking about all this is to follow the lead of British writer Mark Earls and flip the idea on its head: instead of trying to study “influence” by looking at “influencers”, look at the behaviour of the “influenced” instead. Then instead of trying to fit numbers to abstract ideas like “authority” you can think in terms of copying, rejecting, sharing, displaying and be a lot more agnostic about the source of the activity.

In the case of Justin Bieber, this kind of thinking is helpfully in line with my own ideas about pop fandom, which is that a specific star has incredibly little control over their fans. The fan relationship is generally far less compliant and more critical than outsiders imagine. And the focus of fans often isn’t the fan-object, it’s the fandom itself, and the codes and ideas which evolve within it. From the Apple Scruffs to the Bieber hordes, fandom becomes self-organising and self-directed very rapidly, and quite outside the control of the group or person which is notionally its subject. Bieber fandom is interesting because social media allows him a much greater level of one-to-many communication than stars had with fans in the past, but I’d still guess he’s more the engine or battery of the Bieber fan community than any kind of driver. What other fans are doing is, I’d hypothesise, much more important than any top-down communication.

If you like, you could talk about this relationship in terms of “influence” too, but what would be the point, when each relationship of influence – Bieber, or Godin, or Obama – turns out to work on different lines to the other ones, and it’s more interesting to look at them individually rather than try and lump them together?