Can you quantify influence? should you?

Jonny Bentwood has just released his white paper on quantifying-the-impact-of-social-media and while its worth while reading, and there are lots of good ideas in it, I have to say that I fundamentally disagree with the premise. Yes, it's important, I suppose, to some people (Facebook, perhaps?) to know whether Steve Rubel or Robert Scoble is more influential. But ultimately, isn't influence in the eye of the beholder? Personally, because I know Scoble, and love Naked Conversations, AND, I'll admit that I haven't forgiven Edelman for their Wal-Mart Flog, I'll take anything that Rubel says with a larger grain of salt than Scoble.  In reality , neither one can hold a candle to Oprah Winfrey in the eyes of most Americans. Frankly, I would argue that even the ASPCA blog is more influential than Scoble and Rubel combined, given that hundreds of thousands of pet owner relied up on it for pet health and safety tips during the Chinese Pet Recall (besides, who can resist kitties and puppies.)  And, oh by the way, the ASPCA can measure the impact of the blog and their entire PR program in terms of on-line donations. And when it comes down to it, what's more important, as Kami suggests, influence or outcomes?
My point is that this is a  white paper written by and for insiders who have drunk the Kool-Aid. In reality, the only person that really defines influence is your customer. If content (be it on YouTube, a blog, facebook or Twitter) motivates your customers to act, then it's influential. If it doesn't,  who cares? 

Comments

[Technologie Webmaster]

Further - regarding uniformity of influence - alot of network analysis research confirms the phenomenon of "horizon of observability" wherein both perception and direct influence are more or less limited to friends-of-friends.

[Chris Iafolla]

This is the classic PR measurement discussion that discredits the tired clip counting method. For one client, a hit in a niche focused publication might generate 40 leads. An article in the Boston Globe on the other hand might generate 2. Value is measured by how useful it is to the audience.

[Guy Hagen]

Matt S.'s comments are on the mark. There has been forty years of research on social network analysis and social capital prior to the current wave of interest in social media. It is important to remember that linkage analysis best measures what goes through the links - in this case, blog content - and measures that imply influence outside of that are probably less trustworthy. Further - regarding uniformity of influence - alot of network analysis research confirms the phenomenon of "horizon of observability" wherein both perception and direct influence are more or less limited to friends-of-friends. In all probability, the online communities discussing "PR measurement" and "japanese car discussions" are beyond each other's respective horizons.

A good starting point for information on these disciplines is the International Network for Social Network Analysis (http://insna.org/)

[Matt Searles]

I don't know if this is sacrilege to say on a blog dedicated to pr measurement but.. I feel like there's a need to "manage mystery." It's not just a question of numbers, its how you relate to the numbers... one should be cognizant of issues inherent in quantifying stuff, because if you don't.. and just take numbers as truth your marketplace navigation might be less effective then the fellow who cultivates intuition... especially in a space where things move faster then our abilities to quantify... Never mind that there have always been limitations to what we can quantify.

I think you basically need to be like Luck Skywalker. You know that scene with Luke: he's in his X wing about to try and blow up that Death Star? Vader's hot on his tail? What's he do? He turned off his computer and he uses the force! Dig that!

[Alan Chumley]

Until or unless we actually ask those on-line if they were influenced (or apply the relationship methodology), all the indices in the world aren't really measures of influence, but POTENTIAL TO INFLUENCE. It's an assumption. A proxy. And not all such indices are created equal. Some treat bloggers as equally 'influencial' or credible (or one as a contributor to driver of the other) on all topics while others (like Onalytica's Influence Index) try to atleast take that into account. KD, you're an influencial blogger on measurement, but perhaps not on performance modifications for Japanese cars. The indices that treat credibility as equal regardless of topic erk me. Context is everything.

What I DID like about the Edelman paper was the material on the meme starters and mem spreaders as being (again, potentially) more likely to be more credible and (potentially) influencial.

But, we're really talking about Influence Proxy with all these indices.

[Paull Young]

Hear, Hear!

Well said Katie. Often we'll try to gauge influence to predict targets for outreach/communication/relationship development. This will always be necessary.

In my experience, it makes a lot more sense to authentically join the community, engage and let the most influential (for your brand) identify themselves. Standing on the outside and looking in doesn't work, no matter how many numbers you put on it.