Skewing us into tiny pieces

A matter of discretion

And so our tour of the ologies moves from entomology to etymology… come on those at the back, leave the maggots behind we are onto words.

Discrete and DiscreetOne means to separate into distinct groups, and the other is to be careful so as to avoid embarrassment, or to be private.  You can tell which is which because discrete has the two E’s separated by a T (an online image search produces obscure graphs).  Discreet looks like sheet: something which covers and hides, (people saying Shhh, and feminine products).

Mmm, discete uniform distribution.  Niiiice

Mmm, discrete uniform distribution. Niiiice

Not a maggot, no. A discreet ear-piece

Not another maggot, no. A discreet ear-piece

We often talk about discrete groups and behaviours because such divisions are real and can be helpful.

The YouGov Profiler

I mean come ON.   Look at this absolute fucker.

YouGov pen portrait: I mean come ON.
An extra from Dad’s Army?

Anyway, I thought about all this when I saw everyone talking about the Yougov profiler app.  Designed to help marketers develop media plans, here it is, have a go.

Now the good people at Yougov are commendably clear about how it works:

This app does not show the *typical* fan or customer. If it did, most groups would look very similar, and you wouldn’t learn a lot about the specifics of a particular thing


It shows what is *particularly true* about a group. We compare the group to their natural ‘comparison set’ (for example, fans of Downton Abbey compared to anyone who has rated any TV shows) … if something is only true of 1% of the overall population, but is true of 6% of our target group, it might score very highly (and shows you something interesting and true about that group). But it doesn’t mean that it is true of all of them!

(So we see that Man City fans skew towards older women and people who like Myleene Klass are (or were) left-wingers in the North East).

But if groups are genuinely similar, isn’t working with the similarities more valuable than focusing on marginal differences?   Is the 6% particularity more ‘true’ about a population than what it shares with other groups?  It may be slightly less fun than clustering and it’s easier to ignore most of your potential market in favour of your inner core market, but recognising the breadth of your consumers widens your ambitions.

For users of the YouGov app, everything that matters is discrete, while the truth is discreet…

One hates to be churlish about a lovely looking bit of fun from a respected research company.  But this is painfully misleading.

Focussing on what is ‘particularly true’ can be disastrous in TV, which is characterised by broad audiences, watching together. Audiences are rarely discrete, and we should be discreet about such differences that do exist, to avoid the embarrassing marketing clangers caused by over-targeting on ‘fans‘.

And isolating us according to what sets us apart is borderline fascist:

First they came for the hamster owners, and did not speak to me

because I am not a young woman who feels dizzy

Then they came for the fans of First Aid Kit, and did not speak to me

because I am not a young lawyer in Wales

Then they came for the Media Professionals and did not speak to me

because I do not watch only 1-5 hours of TV each week.

Then they came for the people who shop at Lidl, and did not speak to me

because I look nothing like this.

Then they came for me, and missed the fact that I’m

a hamster-owning, Lidl shopping, fan of First Aid Kit working in media.



The topic of pigeon-holing, and writing off most of your target market is echoed in the words to this beautiful song.  Imagine it’s sung by a (potential) consumer for your product, but who falls outside the particularity of your target, so you ignore them.

I could’ve been so many things
But it would never be enough for you
I was the one you counted on
But I was never the one for you
Now I know
I lost you a long time ago