How things spread

Posted by on Apr 26, 2008 in Uncategorized | 6 Comments


Been thinking about this question quite a lot recently so here’s the first in a series of posts to clariy things – one thing at a time.

Ok, the thing is that we assume that it’s what m>we do – in marketing or management or government – that changes stuff out there in the world: that gets a behaviour or an idea to spread through a population.

Of course, this is only natural given that’s what we get paid for – which politician wants to stand up and say “crime is down – not that our policies have much to do with it”; it’s largely due to factors beyond our control.
Which marketer is going to admit that the growth of their brand has bugger all to do with marketing’s actions (premium vodkas: discuss). Which manager with a KPI to hit….

The truth is that it’s all about them out there: most behaviours and ideas spread through populations because of what the members of the populations do or think or say in response to each other. That’s why most big social trends surprise us – why as Freakonomics pointed out, the “broken window” strategy to clean up NYC is probably less important in explaining the 90s reduction in crime in the city than the demographic changes kicking in.

Put simply: as far as how stuff spreads, it’s more Pullyou than Pushme (to bastardise the old Dr Doolittles). Folk do what they see around them; believe what their peers do – whatever their individual brains tell them. Our attempts to exert ‘exogenous’ (extra-system) influence is always going to be much less important than the ‘endogenous’ (intra-system) factors that shape the propagation of an idea or behaviour through a population.

BTW Think this is what Cluetrain really meant about markets being like conversations

If this is right, then strategy thinking needs to be rooted more in how the underlying mechanics of propagation works (human-human emulation within a given system etc) than in why individual folk do what they do.

Put another way: forget doing stuff TO folk; do stuff WITH them


  1. neilperkin
    April 26, 2008

    Nice post Mark. Reminds of that whole thing about how you can’t ‘create’ a community, and nor should you attempt to, but instead think about how you can help that community do what it wants to do.

  2. john.griffiths7
    April 26, 2008

    one of the gems I got out of the recent WOM how to workshop was the idea that you can use research to find out the shareability of an idea-this is about process not content – and process is the bit that normally gets discarded in the research process

  3. Mike Tyldesley
    April 27, 2008

    Mark, I haven’t posted in a while, but I’m still reading and getting a lot from the blog. This post is incredibly interesting ingiving perspective to some work I’m involved in with the Rugby (League) club I support.

  4. david cushman
    April 27, 2008

    Mark, I get that this informs the ‘how’ of strategy, but less the ‘what’. Hope you’ll address that in the rest of the series – and when we meet this week. cheers dc

  5. david cushman
    April 27, 2008

    One further thought to throw into the mix, re process ‘shareability of an idea’ (john.griffiths7)… to paraphrase Clay Shirky, if the transaction cost of group activity (of groups forming) has fallen through the floor, perhaps all ideas become as shareable as all others? The network allows any group of any and all purposes to form in real time on a planetwide scale. Does that take us back to having a good idea to share in the first place?
    That’s kind of my ‘what’ rather than ‘how’ point in the previous post.

  6. Andy
    April 28, 2008

    Mark, thanks for opening up such a rich vein for exploration. It has set off a lot of loosely connected thoughts.
    There seems to be useful overlap here with the work Dave Snowden is doing /has done looking at actions under conditions of differing uncertainty/non-linearity (cynefin framework). Have just watched the WOM vids and the options strategy Scott talked about is analogous to Snowden’s strategy of construction/deployment of “probes” in a complex environment.
    Re: David C’s last comment… the point is that we don’t know in advance what might constitute a “good idea”… the herd decides this. There might be some things we can do to increase noticability etc. The analogy here would be with the idea of an attractor. So put a number of ideas out and see what happens (this is the “what”?). The more uncertain you are or the environment is, or potentially the more chaotic it is, the less it matters what you do and the more it matters that you simply act so you get feedback. For better or for worse.
    If you know in advance what constitutes a good idea, you’d just implement it (ah the benefit of perfect foresight).
    I agree with your comment about doing stuff with them (really there’s no choice anyway… the distinction is how we view it as marketers to give the illusion of control). To this end when we enter the conversation we have as much chance as anyone (?) to reinforce or dampen any attractors that are forming.
    Further to the experiments you/Scott and Duncan were proposing I do wonder about repeatability – under conditions of complexity, context matters… just because it worked last time, there are no guarantees this time even given the same starting conditions. Or is there something here about human nature which is more stable?