Last week, CMSWire’s Chelsi Nakano wrote a short review titled “5 Business Tools for Measuring Social Influence Online.” It caught my eye because, in addition to my web analytics practice at Web Analytics Demystified, I am the CEO and Founder at one of the measurement services Chelsi covered, Twitalyzer. While I think Chelsi did a good job covering some of the more well-known technologies in the space — ourselves, Klout, PeerIndex, Hubspot and Empire Avenue — the interesting question Chelsi didn’t ask was whether social influence should be measured and how those measures should be used.
What Does “Social Influence” Even Mean?
Measuring influence is difficult at best, given that it is one of those “I know it when I see it” phenomena. In the real world (i.e., offline), social influence is often easy to recognize — for example when a politician makes an impassioned speech, or when an athlete or actor starts a new fashion trend. While the barrier to influence is high — the politicians, actors and athletes need to have a platform to speak from — the result is usually clear. The votes come in and the clothes sell out, allowing us to collectively agree that the politician, actor or athlete has influenced the collective behavior of whatever network or group they are targeting.
An interesting challenge arises however, in the online world powered by Facebook, Twitter and YouTube. Whereas in the real world, you actually need to be someone or do something to be given a platform to influence from, thanks to social media these same barriers don’t exist. Here we all make our own networks, usually organically, and with audience squarely in hand we set out to influence our individual networks.
How Do You Measure Social Influence?
Of course, since social media exists in the most measurable of all media, it was inevitable that geeks like me, Joe Fernandez at Klout and others would step in and attempt to quantify social influence on the Internet. All of the companies that Chelsi cited have come up with some strategy to assign a numerical value to the ability of any individual to influence other people.
At Twitalyzer we have nearly 30 measures, five of which are primarily focused on measuring social influence (“influence”, “impact”, “engagement”, “generosity” and “clout’.) Like Klout, PeerIndex and the others, we calculate these measures in order to help shed some small light on otherwise anonymous individuals, something that is especially useful if you are a business owner or marketer looking for brand champions, though-leaders and other potential allies in social media.
The Emerging Problem with Social Influence
The problem with these measures — and I certainly am not excluding Twitalyzer despite my ownership of the company — is that some people are taking them entirely too far. I detailed this in a blog post I wrote last week that was prompted by a conversation with Shel Israel, author of “Naked Conversations” and “Twitterville.” During our conversation, Shel pointed out that he knows a consultant who was passed over for work because he had a “low Twitalyzer score.”
Suffice to say I was shocked and dismayed to learn this.
Again, despite my financial interest in Twitalyzer’s success, I think that anyone using our data, Klout, PeerIndex or any other online measure of “influence” to make a real world decision is a fool and making a huge mistake. All of these measures, regardless of what they purport to do, measure nothing more than an individual’s use of social media. Moreover, unless you are clearly an influencer offline (see Lady Gaga, Justin Bieber, Barack Obama, etc.) your ability to influence online behavior has little or no impact to your influence in the real world.
You can use online measures of social influence to make decisions relevant to the online world. Trust me, I love it when our customers tell us they used Twitalyzer to identify bloggers and other individuals who successfully helped them spread their message. But I strongly encourage readers to be careful not to place too much faith and too much import into these metrics ability to understand honest too goodness, flesh and blood relationships and interactions.
This is a conversation that is going to go on for some time, and one that nearly everyone has a fairly strong opinion about. I would love to hear your thoughts, either here at CMSWire or over on my original blog post.
About the Author
Eric T. Peterson is the founder of Web Analytics Demystified and has worked in web analytics for over 10 years as a practitioner, consultant, and analyst. He is the author of three best-selling , Web Analytics Demystified, Web Site Measurement Hacks and The Big Book of Key Performance Indicators, and one of the most widely read web analytics writers at www.webanalyticsdemystified.com.