Duncan vs. The Influentials
Much has been written about how the influential few [an elite 10%] tells the rest of us what to buy, how to vote, etc. There is a book on the topic naturally named The Influentials. A tiny cadre of highly connected elites influencing the rest of us was a key theme in Malcolm Gladwell's The Tipping Point.
Most people would look at the social graph below and say Diane is the influential in this sub-group [all of this group's connections and contacts are not shown]. She has local reach, but her message gets nowhere without the help of her network. Influence needs many connected people to spread -- not just the highly connected. Heather, Fernando or Garth all need to be in a cooperative mood for Diane's message to travel.
In social network analysis, a boundary spanner is someone who "crosses the chasm" between groups/clusters. They are not often highly connected. In the above network Fernando, Garth and Heather are all boundary spanners. They may not be influential, but they need to be ready to accept the message/trend/idea if it is going to make the jump out of their local domain and travel further. Otherwise the innovation/idea bounces around and dies in a cul de sac.
Duncan Watts, at Columbia and Yahoo! Research, has been slowly dismantling "The Influentials" theory. This Fast Company article is a good overview of his argument. He basically says that it is not the elite few that matter but the connected many and they have to be ready to be influenced. I'm with Duncan.
Most people would look at the social graph below and say Diane is the influential in this sub-group [all of this group's connections and contacts are not shown]. She has local reach, but her message gets nowhere without the help of her network. Influence needs many connected people to spread -- not just the highly connected. Heather, Fernando or Garth all need to be in a cooperative mood for Diane's message to travel.
In social network analysis, a boundary spanner is someone who "crosses the chasm" between groups/clusters. They are not often highly connected. In the above network Fernando, Garth and Heather are all boundary spanners. They may not be influential, but they need to be ready to accept the message/trend/idea if it is going to make the jump out of their local domain and travel further. Otherwise the innovation/idea bounces around and dies in a cul de sac.
Duncan Watts, at Columbia and Yahoo! Research, has been slowly dismantling "The Influentials" theory. This Fast Company article is a good overview of his argument. He basically says that it is not the elite few that matter but the connected many and they have to be ready to be influenced. I'm with Duncan.
posted by Valdis at
3:28 PM
4 Comments:
Matt Ingram scratches the surface in a related post ... "influence isnt all or nothing". http://www.mathewingram.com/work/2008/01/28/influence-isnt-all-or-nothing
Brian Uzzi points out the various values of heterophilic and homophilic connections (i.e. balance is best): http://www.kellogg.northwestern.edu/faculty/uzzi/ftp/research.html
I am with Duncan in pointing out that there is no group of bloggers or politicos that consistently monopolize the smoking gun of influence. I understand that polarizing his view against Gladwell's may promote more attention and therefore lead to better understanding.
Each diffusion is unique and depends on the fluctations and prevalence of related environmental cues in the idea habitats being examined (Berger & Heath, http://marketing.wharton.upenn.edu/Marketing_Content_Management/Marketing_files/Publication_Files/Idea%20habitats.pdf )
Doesn't Gladwell start out by suggesting a marketer should take steps to identify the specific group of influencers that might work to advance their objectives?
If you have an agenda to advance or a product to market, can you not increase your likely hood of success by targeting a group of likely early adoptors and then be vigilent for the boundary spanners from that group who will lead you into the early mainstream market?
In other words, is there any merit to taking a strategy that targets a group of early adoptors and then zeros in on supporting the influencers within this group to obtain a "tipping point" when the idea or product begins to cross Moore's "chasm" and being spread by larger population segments?
Yes - the question of whether the larger population will accept the advances of your group of influencers will depend on their preconceptions as well as the credibility of your influencers.
Perhaps Duncan provokes us on a more stategic level (product/idea positioning and design or as Seth Godin might enthuse, "find products for your customers, instead of customers for your products http://sethgodin.typepad.com/seths_blog/2008/01/tribal-manageme.html) and Gladwell is picking up on the tactical dynamics (marketing communications)?
What do you think?
By
Michael Cayley, at 2/11/2008 10:40 AM
Great post! Michael's points are dead center, and Valdis, I think your post emphasises a very important distinction in discussions of "influence". I continue the discussion in context of social media in my post "Influencers, Gatekeepers and Average Joes": http://innovationinsight.com/blog/2008/influencers-gatekeepers-and-average-joes/trackback/
By
Guy Hagen, at 2/11/2008 1:05 PM
I seem to remember reading about percolation/social contagion in Six Degrees...where it discussed how a certain percentage of one's connections have to have adopted an idea in order for it to appeal to a given individual.
So, in the network that you showed, if you set the necessary threshhold to 50% of your friends have to adopt an idea before you adopted it, you might try to convert Carol and Beverly. They would convert Andre, at which point Diane would be susceptible.
At that point, it could spread to the rest of the network. But Diane herself could not do that job without the support of the others, and it might actually be inefficient to try and influence Diane directly(in the real world she might be the target of many marketers trying to convert influentials, and would have a high resistance to such messages).
Perhaps in real-world situations someone like Fernando might reject a message from a single person with a rationalization like: "Well, buying X is good for Diane but not me because she is so popular, unlike me". But if Carol and Andre also told Fernando how cool X was, he might listen more readily.
Interestingly, I think this is how it played out for me when I decided to switch from a Windows PC to a Macintosh. Most people around me for a while had been PC users, but I started working at a company where most people I interacted with used Macintosh. Even with ads telling me how great Macs are, or Influentials proclaiming their superiority, it took conversations with people in my work network to make me feel comfortable in buying one for myself.
By
synchrono.us, at 2/12/2008 8:06 PM
Wow, imagine that--the influenced might have some will of their own. It seems to me that we need both sides of the situation. Influencers influence....those willing to be influenced. Now, what gets interesting to me is the qualities that make the influencer able to influence and what are the influenced looking for when making the decision to be influenced? The characteristics that convert an influencer might have more to do with the quality of the thing they will be promoting...but the influenced might be more susceptible to the expertise, credibility, quanitity, or even reciprocity of the influencer. I have been enjoying Cialdini's work on this in "Influence: the Psychology of Persuasion".
By
Nurture Girl, at 2/22/2008 5:52 PM
Post a Comment
Links to this post:
Create a Link
<< Home