I’ve written before about my doubts around the concept of social influence. Instinctively I was and am of the view that the measurement or ideas of online social influence are clumsy at best and misleading at worst. I had the pleasure of meeting John Sumser at #HRevolution a couple of weekends ago and whilst I really enjoyed our, albeit brief, conversation, I’m not convinced by the model that is used by HR Examiner, nor indeed any of the suggested measures or rankings used by others (I daren’t even mutter the name beginning with K).
Last week I was fortunate to spend some time listening and discussing social networks with Nicholas Christakis. Christakis is somewhat of an expert on social networks – you can see his TED lecture here – his focus is on real social networks…the old sort, people who talk to people, like we used to do.
Listening to the work that he has undertaken I started to pull together my thinking a little more to try and explain my instinctive discomfort.
The HR Examiner model looks at three main factors,
Reach: A measure of the audience size (number of eyeballs) for each individual. Traffic.
Relevance: The degree to which content associated with the individual matches a cloud of keywords prepared for the analysis
Resonance: The number of mentions, inbound links and participation found for each individual
Which sounds relative sensible. Until you start to really unpick the concept of “influence”,
“the capacity to have an effect on the character, development, or behaviour of someone or something” – Oxford English Dictionary
And the critical word is effect. There needs to be an effect, in other words some sort of change, which isn’t measured by any of the online registers of influence. The argument goes that the “resonance” goes some way to addressing this. But does it?
Christakis mentioned a really interesting real life example that had happened to him. He is co-author of a book called “Connected: The Surprising Power of Our Social Networks & How They Shape Our Lives” and last year a link to the Amazon page for the book was tweeted by Alyssa Milano (no I didn’t have a clue either) who has 1.5m followers on Twitter. The tweet went on to be re-tweeted nearly 2m times.
Being scientists, Christakis and his co-author went and looked at their sales through Amazon in the following period.
There was no discernible change AT ALL.
But that is because Alyssa is a TV personality, right? And therefore there is no relevance. So the authors asked Tim O’Reilly to tweet the link to his 1.5m followers. Which he did and which was then retweeted and retweeted. When the authors checked back in on the sales,
There was no discernible change AT ALL.
The story goes that hearing about this, Susannah Fox (social network researcher and author) tweeted the link to her 5,000 followers believing them to be a highly relevant demographic.
And guess what? There was no discernible change to sales at all.
Both O’Reilly and Fox had Reach, Relevance and Resonance in abundance. But this hadn’t led to any action, they had failed to actually “influence” anyone. Why is this? I’d be lying if I said that there were any concrete answers, but there were a couple of things that we discussed that I think are at the root of the issue.
Their research has indicated that there is a big difference between what I would call “contacts” and “friends” on social networks (they specifically looked at Facebook). There is a big personal factor on people’s propensity to be influenced. And that of course makes sense, we are more likely to take action based on the words or advice of people we like, trust and want to be like, than people we know nothing or little about. The “influencers” may have a lot of followers, but how many of them are just contacts and how many are in a real relationship with them?
Secondly, Christakis talked a bit about the “influencable” suggesting that we focus too much on trying to identify the shepherds and not enough time considering the sheep. Influence isn’t a purely passive game of pushing out information, it requires a certain level of complicity, on either a conscious or subconscious level. A lot of the way in which we look at online influence measures propagation alone, or as Christakis puts it,
“I’m not saying that Twitter is useless, but I think that the ability of Twitter to disseminate information is different from its ability to influence behaviour.”
So are we dealing with real influencers or are we dealing with a relatively select group of individuals churning about the same information, tweeting, retweeting and commenting about one another in some sort of mutual beneficial self-fulfilling prophecy?
Well I am absolutely convinced that a number of people on these lists are doing good work and doing it for the right reasons. However, there are a lot of people doing that who don’t spend their time tweeting, retweeting and connecting to people on LinkedIn. Does that make them any more or less influential? And the simple fact is that we don’t know.
And why don’t we know? Because we simply can’t measure it without undertaking a breathtaking longitudinal study. Social media and the internet allow a lot of people to profess expertise and to build profile without any just cause or track record. The risk of online measures like these are that they could, unwittingly, legitimise this.
I don’t want to beat up on a particular measure or organisation, that isn’t my intention, but I will be viewing these accreditations with a huge dose “scientific and behavioural research” salt, until they can prove otherwise. I’d advise you to do the same.