Sunday, August 01, 2010

The folly of categorisation...

Harsh, but fair.

In recent years, I've become a bit of a muso, as the odd post on this blog (and how I spent my student loan) should testify.

And, in no other part of my life do I get as much pleasure as recommending a new band, or an old band that no-one else knows.

There's the old horse chestnut when you have to describe a band to someone else, and compare them to others. I'd argue that there are bands or artists who take such a right turn from their usual sound (though I don't like him, witness Plan B recently, or say Radiohead for Kid A) that they defy categorisation.

This sort of thing is why I worry a bit when people apply arbitrary labels to people to explain their behaviour. An 'early adopter' for one product doesn't necessary apply to the next one they release. I love the iPod, but would I buy a Mac? No, and there are a variety of reasons. I would consider Nike for running shoes, but fashion trainers? Not a chance.

Segmentation is fine when it works for broad behavioural patterns, but the whole Gladwell bell curve attempt by agencies to neatly fit people into an assumptive model, or to assume buying patterns somehow have a rational pattern is bullshit.

Much as I find him to be a grumpy bastard when he writes, this is something I cede to Taleb. People are too chaotic, and life is too random, to assume that the most middle of the road strategies are going to work. Why not do a combination of the safe and highly dangerous when planning or executing campaigns?

Middle of the road means your market share will atrophy. Grouping consumers as early adopters means your values will parallel theirs; witness brands which chase an ideal too strongly; one which has gone out of fashion (say most mobile phone brands and having an eye on the future) - or those which succeed by re-harnessing an ideal which has come back in (say, Old Spice or Hovis).

Daily, people defy audience segmentation. So why do we bother? Increasingly, it looks like something which results in jobs for the boys; a lazy back up plan for weak-willed Marketing directors.

I'd far rather be a brand which did the basics brilliantly and hedged its bets on consumer behaviour, rather than executing a strategy which has been passed around so many people that it now bears no resemblance to what was first presented.

Given these trading conditions, strategies either have to be so, so basic (I'm thinking of a certain jeans brand's recent work) as to seemingly insult the intelligence of its audience and not really say anything, or contain a lot of mixed messages which don't DO anything.

We talk in hushed tones about a 'purpose idea' or a 'brand ideal', but all of this is bullshit if it relies on the sort of Stone Age segmentation which a lot of marketers seem to be so fond of. People just aren't a brand character; they have more interesting little niches or jagged edges - it's those which'll make money going forward, those fascinating gaming inspired Easter Eggs (like Google's Pacman display, or Dole's approach to labelling) which tell you more about the people who are going to be your consumer for the next twenty years, rather than an empty current figure.

In fact, gauging the lifetime value of a consumer is interesting these days. A Facebook 'fan' is obviously not a reliable metric here. As discussed on twitter, there's quite a gap between being a fan and being an advocate, someone who will keep on buying.

A personal example - I love Adidas trainers. I like the style, I admire the Predator connection, and love their golf clobber. Yet, I think their marketing (compared to Nike) is often a bit amateur hour. I wouldn't favourite their stuff on Facebook, but give me some money off some Stan Smiths, and I would buy. Nike, I'd love the thinking behind the work, but would I buy their trainers? Not a chance.

Digital metrics are great. I think it's wonderful to be able to gauge the sentiment behind work, and see how well it's been received online. But would I rely on them to knock out a segmentation, or be able to tell how easily my product would fly off the shelves? Not a chance. Would I use them to figure out how to place my budget, and how much of it is for straight promotional activity and how much of it is for more chaotic activities? Damn right.
Google Analytics Alternative

Enter your email address:

Delivered by FeedBurner