European tech conferences don’t often see flare-ups of controversy, but the first London edition of the long-running Le Web conference managed it. Last Tuesday, Joe Fernandez was interviewed on stage by Alexia Tsotsis of Techcrunch – and made a spectacularly poor fist of defending his online influence-measuring product Klout live on stage.

Alexia asks if she were too hard on Joe. I think that’s the wrong question – or, at least, not the most interesting question. I’m more interested in why so many people, both those at the conference, and those watching over the internet, were so keen to see Klout fail.

There exists an inherent tension between companies with marketing structures rooted in traditional paradigms wanting to be able to identify influencers, and people’s desire to been seen as something more than a simple number.

The roots of this are clear – the marketing industry has lost its clear guidelines as to influence and the people who wield it. With the narrowband publishing channels of magazines and television being steadily replaced by the broadband publishing environment  internet, there are no longer a handy, limited group of influencers for marketeers to target. Who is influential in this noisy world of blogs and social media? Whom do I focus on to get my message across? Journalists’ influence was easily measure in part by their own reputation, but mainly by the circulations of the titles they wrote for, as more than one journalist has discovered to their shock when they left a magazine and found their influence destroyed. It’s traditionally been much, much harder to judge the influence of a blog, and the last major and generally well-regarded attempt – Technorati – switched long ago to being an advertising network. Back in the day, though, it did much the same thing that Klout and its ilk are doing now – assessing something’s influence by rating it with a number.

Klout, and other services which claim to be offering some measure of influence, including PeerIndex and Kred, have focused on the much more visible data we can see in social networks, rather than link-based assessment of blogs. (It’s interesting to note that none of them have made any significant attempt to draw people’s blogs into their influence rating…) They use that in an algorithmic way to assign a value to a person. And that’s where the problem begins. The moment you start stamping people with a numeric assessment of their social worth, you’re going to create controversy. For some, sure, it will spark their competitive edge, and they’ll enjoy the process of trying to make that number get bigger. For others, it will being an unpleasant reminder of school exams or work performance assessments drawn into their personal lives.

For a thinking audience – and thankfully digital conference still attract thinking audiences – the idea that something as complex, variable and human as social influence can be summed up in a single number – one with no clear picture of how it’s generated – will be distasteful. The online reaction to the interview proved that. Fernandez made some sensible noises about making changes to the algorithm to take more account of real world influence – but what I think really needs to happen is for these services to be much more explicit that their services are only one element of judging and understanding the vagaries of online discourse and influence.

For anybody hoping to achieve a mature understanding of influence, a numerical score like this can be a useful data point – but only one amongst so many more.