A new lore of averages: what Clay Shirky and the Coney Island Mermaid Parade can teach us about defining target audiences

insight after insight from Clay Shirky's Here Comes Everybody.  the above chart is copied from chapter five which covers collaborative production. it shows contributors to the Coney Island Mermaid Parade Flickr site ranked by the number of photos they contributed.  a couple of users contributed the most whilst the most users contributed only a little.  Shirky observes that:

"...the imbalance is the same shape across a huge number of different kinds of behaviours.  a graph of the distribution of tags on Flickr is the same shape as the graph of readers-per-weblog and contributions-per-user to Wikipedia.  the general form of a power law distribution appears in social settings when some set of items - users, pictures, tags - is ranked by frequency of occurrence"

that there's a massive imbalance between people who contribute in collaborative projects we know.  but its something that we don't often enough plan for in a media world where we increasingly ask (the audience formerly known as) consumers to user-generate and co-create on our behalf.  Shirky goes on to point out that:

"...the imbalance drives large social systems rather than damaging them.  fewer than two percent of Wikipedia users ever contribute, yet that us enough to create profound value for millions of users ... the spontaneous division of labour driving Wikipedia wouldn't be possible if there were concern for reducing inequality rather than limiting it ... large social systems cannot be understood as a simple aggregation of the behaviour of some nonexistent 'average' user"

and there you have it.  he said it.  there's no such thing as the average user.  we all know this, and yet we still struggle to capture the targets for our advertising campaigns in neat tangible soundbites.  the demographics of old have (thankfully) long gone, but whilst they're been replaced by more contemporary means - attitudinal or usage based targeting - our one-dimensional thinking too often remains...

we are still, by and large, expected to think of and present 'one' target audience.  an 'averaged' person or group based on some attribute of attributes that are most relevant to the brief.  but look again at where the mean 'average' sits in the above chart... it not only fails to capture the few individuals who would be super-involved in what we have to say or ask them to do, but massively over-estimates the extent to which most people will commit attention to our branded projects.

we need a new lore of averages for our targeting-think.

when we describe target audiences we should be thinking of them as sitting along the above spectrum.  how do we plan on one hand for the very few but valuable super-attention givers from whom a lot of the effectiveness of the media investment will derive?  whilst on the other hand plan for the 'mode' individuals, the vast majority who will contribute the smallest amount of attention to what we have to say?

this spectrum, this logarithmic curve of attention, exists at whatever level you aggregate.  be it a population, or an age range, or any segment no matter how - whether attitudinally or behaviourally - it is defined.  there is no average user, no average consumer, no average contributor, co-creator, or co-collaborator.  let's stop kidding ourselves and clients otherwise.

Comments

[Pete]

Interesting post, 'Here Comes Everybody' is on my ever-growing list of books to read.

Like you say, there is definitely a need to take into account the imbalance between how involved different types of people will be.

I think it is being considered more and more in social media campaigns where they get bloggers or vloggers to take part in some kind of experience that usually involves the brands product in some way (Ford Fiesta Movement is quite a good example http://www.fiestamovement.com/ ). The subscribers/readers of these people taking part also come along for the ride; so the companies are targeting the few super-involved knowing they will also reach a wider more passive audience as a result.

Though in examples like this are the mode being ignored? There's either a high level of participation required or you're just an observer. Maybe there should be smaller ways people can take part so however deeply you want to get involved there is something for you. For example in the Ford Fiesta Movement they could have got people to suggest missions for the 'agents' to do rather than Ford just setting the missions themselves, which seems to be the case. I think little things like that could help encourage more people to take part.