CNET has conducted a study of 'influencers' amongst users of the websites in its network.
Entitled "The Influencer Study from CNET Networks: Challenging Perceptions," CNET decided to test the Influentials theory that it notes many marketers have accepted:
"Conventional wisdom holds that influence is not widespread, but is the domain of a few high profile individuals who have a deep expertise in certain subject matters and advise the rest of the public on matters of choice."
"They are further assumed to be highly connected, with very large networks of people to whom they can impart their knowledge. This notion has informally shaped media and marketing models in which influence is visually depicted as a pyramid, with a few highly influential individuals concentrated over a mass of others who consume but do not advise."
CNET decided to put this conventional wisdom to the test using a three-party study that included quantitative research, qualitative research and a behavioral analysis.
When all was said and done, CNET's study resulted in an overall conclusion that isn't that far off from the one Duncan Watts came to after he conducted his research:
"The flow of information isn’t coming just from a small group of connected individuals at the top."
The details, of course, are what make this conclusion interesting. The three that I find to be most relevant:
- "Data suggests that the 1-to-10 or pyramid model, where influence comes from a few highly-connected people who advise the unconnected masses, is inaccurate, and points instead to the significant potential of the moderately-connected majority. So, the true shape of influence may be better characterized as a diamond, not a pyramid. Focusing only on the highly-connected few misses a huge opportunity to communicate on a large scale with the bulk of the population."
- "Conventional wisdom holds that the highly influential few are topic experts, and share their deep expertise with the masses in part to demonstrate their knowledge. However, our research finds that influencers are primarily motivated by a desire to help others. People like to be needed and valued, and influencers derive a sense of self-worth and validation from giving good advice. They aren’t simply blasting e-mails to their entire address book – these influencers are taking the time to seek out and customize information they believe will be relevant to specific individuals within their network."
- "The study reveals that for information to be considered valuable, it must be both unique and trusted. For example, an AP headline might be trustworthy, but it isn’t unique – one can find it in dozens of places. And information on an amateur fan site might be highly unique, but not necessarily trustworthy. Sources that provide unique and trusted information will be referred to again and again."
- As I have noted, mass-marketing is still important because you simply can't be sure which consumers are going to take your message and run with it. Simply trying to target consumers who you think are most influential is probably not the best strategy - targeting a wider swath of consumers probably is. After all, according to Duncan Watts, "If society is ready to embrace a trend, almost anyone can start one--and if it isn't, then almost no one can." If Watts is right, limiting the number of consumers you target may very well be limiting the potential of your campaign to succeed.
- The notion that "Influentials" act as brand zombies who blast recommendations to their contacts is naive. As I've argued before, the power of word-of-mouth is derived from the way it's delivered. Those seeking to artificially create word-of-mouth and to leverage it formulaically miss the point - recommendations are inherently powerful because we aren't bombarded by them and when we do receive them, they're typically delivered in a personal fashion from somebody who consciously believes they're relevant to us.
- Trust is important, but having a message that is relevant and unique is too. Trying to get consumers to promulgate a message about laundry detergent, for instance, probably just isn't going to cut it because there's nothing truly valuable about the message, even if the source of the message is trusted.
It's necessary to note that CNET is not completely unbiased because it does derive a significant amount of revenue from advertising campaigns that are probably more mass-market than "social."
But, as is becoming typical, the studies with quantitative data cast doubt on the unnuanced concept of Influentials put forth by individuals who typically have no quantitative data and who fail to explain the actual dynamics of the influence they think exists. In other words, doubters of Influentials typically have actual research while proponents typically have faulty inductive reasoning based on observations and anecdotes.
Of course, none of this will stop Influentials proponents from continuing to promote a theory that just doesn't stand up to scrutiny. But that's to be expected because, ironically, these people apparently just can't be influenced – no matter how compelling the data demonstrating that their worldview is, at best, flawed and, at worst, completely false.
Looks like I’ll have to find the most powerful Mavens and Salesmen to convince them.