Influentials not as influential?
Posted on | October 12, 2008 |
This is a translation of a post from my previous blog in swedish. After WOMMAs release of the Influencer Handbook, it might be good to remember that Influencer Marketing isn’t a “cure all”. So, let’s get to it:
When it comes to starting trends, we’ve seen a revival of the two-step flow of communication, thanks to books like Malcolm Gladwell’s The Tipping Point in 2000 and Jon Berry and Ed Keller’s book The Influentials from 2003. The idea of “Influentials”, originally proposed in the 50’s by Elihu Katz and Paul Lazarsfeld, has been revived. The idea proposes the “two-step model” of information where, if you manage to reach the few influential people within a community, these few will influence the masses and the tipping point will be reached.
However, Duncan Watts, a network-theory scientist from Columbia University, has recently challenged this theory in an article in Fast Company. After analyzing e-mail patterns he has found that highly connected people are not really the social hubs we expected.
Watts created a computer simulated society to test, who actually manages to create trends. The conclusions were that in the large majority of cases, the spread started with an average Joe, but when an influential started it, it spread much further. Even when the Influentials had 40 times the reach of a normal person, you couldn’t be sure they could kick-start a trend. Why? Watts believes it depends on how susceptible the society is overall to the trend.
So, the conclusion basically is that there are people that are more influential, but they still cannot tip the trend if the society as a whole isn’t ready. Gladwell comments:
“I think that all books like The Tipping Point or articles by academics can ever do is uncover a little piece of the bigger picture, and one day–when we put all those pieces together–maybe we’ll have a shot at the truth.”
So, what to do then? Watts has been digging into this too. He has researched the idea of “Big Seed Marketing” – built on two insights: Cascades require word-of-mouth effects, but since you cannot know who is going to start the trend, you should try to reach as broad a market as possible in the first stages. Neither of the tests Watts ran could be described as viral hits, but he usually managed to double the spread compared to the initial recipients. The missing link, as I see it, would be good content that truly fits the environment or community you want to reach.
As a final conclusion: To start a trend, you need to reach out on a large scale, to average Joes as well as Influentials. But, to get as big a spread as possible, you should spend some effort trying to target the influentials specifically. And key to get a really wide spread is that the community or the environments you wish to reach, are receptive to the idea.
Comments
6 Responses to “Influentials not as influential?”
Leave a Reply
October 12th, 2008 @ 9:20 pm
[...] is a follow up (also a translation from my post in swedish) on the last post - Influentials not as influential, where we looked at the “two step model” of information and the belief in Influentials [...]
October 16th, 2008 @ 9:22 pm
[...] means that it is the first person, which means that it could be anyone - influential or average Joe doesn’t matter. Spike continues: “If that conversation is filled with honesty, [...]
January 12th, 2009 @ 10:11 pm
[...] toast? - that influentials are not as important as we’ve believed (I wrote a take on it too - Influentials not as influential?). So, many connections doesn’t necessarily mean that a message will go viral, but once it [...]
July 26th, 2009 @ 9:19 pm
[...] addition, I would like to add that there is probably another factor at play. As mentioned in Influentials not as influential, and Context and virality, whether the environment/society is ready or not is key for spreading an [...]
September 6th, 2009 @ 8:54 pm
[...] just as Duncan Watts says, context, or how receptive a society as a whole is to an idea, is extremely important. I assume this too will affect if people will pass something on, as well as [...]
April 14th, 2010 @ 10:15 pm
[...] From research by Duncan Watts, we have learnt that influentials aren’t that influential. [...]