Influence, the influentials and the influenced

Posted by on Feb 4, 2010 in behaviour change, Copying, Science, Weblogs | 3 Comments

Nice musings by chum David C and Stowe Boyd around this study that suggests it's the not the number of connections you have that determines your influence, but rather the nature of those connections – your "betweenness" as Stowe calls it:

"It is not your follower count, or who you follow, per se. But, instead,
do you have short paths into other social scenes, both incoming and
outgoing? That is the deep structure of being truly connected: bridging
over different social scenes, acting as a conduit, a vector, a filter
and amplifier for ideas good and bad, the best insights, and deadly
viruses." [SB]

But David makes this important point:

"Influence is fluid. It resides less in the node and more in the
interactions between the nodes. It is the interactions which change the
state of the group, not a change in the condition of the nodes (think
water H2O molecules and ice, water, steam…

…This means that giving interesting things to people to do together –
bringing them together around things they care about (through shared
purpose), to act on those things, has more value than spotting the
influencer and giving them some sort of message you expect them to go
off and influence others with"

Which prompts my sixpenceworth:

Influence is not something done by certain people to other people (as for example the pic above from New Scientist might suggest) it's the result of those people we call The Influenced doing something in response to those we call Influential.

As Duncan Watts and Matt Salganik noted in their important music-download paper, it's more about the readiness of a population to adopt a behaviour than the behaviour of specific individuals.

I.e. The Influenced do the heavy lifting of Influence, not the Influentials – and the fact that Influence is often mutual i.e. we all (well-connected or not) take cues from each other – that's one important reason why the search for the Influentials is so prone to misunderstanding and dead-ends.

As soon as we grasp this, going about understanding and harnessing influence begins to be a little more sensible…


  1. Jan Van den Bergh
    February 7, 2010

    The strange debate continues. It’s a chicken and egg discussion. You need influentials to influence the influenced. But of course you also need people who have “the readiness to adopt a behavior”. Influentials are only 15% of the citizens and consumers. The other 85% often let themselves influence by these opinion leaders.
    Without the 15% the other 85% wouldn’t exist and vice versa.

  2. Mark Earls
    February 7, 2010

    Got to say I disagree. It’s not a chicken and egg issue.
    The most common underlying structure of social networks is not hub-and-spoke (as the influentials hypothesis suggests) but either small world or effectively random and mutual.
    Also, seems to me you still want insist on importing “influence-as-something-that’s-done-to-people” into our descriptions of how stuff spreads? Why not accept/consider it from the other way round?

  3. Leighh
    July 6, 2010

    This might sound odd but I read this study this morning about divorce being contagious ( ) and it occurred to me that what seems to be missing from the research is ‘context’.
    Influence is often contextual and yet, very little goes on from a marketing perspective to look at that underlying context bc it’s too much manual labour.
    Paying a fitness blogger with traffic to reco my product is far easier than finding the 100 influence health needles in my 100K chubby and guilty hay stack.
    And worse, even if i find a way to reach and engage those 100 influencers? If context turns out to be king, they might not work for my next go around. Clearly not sustainable.