ELECTION AUTOPSY: Why last night’s ‘shock’ result proves the oldest adage of market research


I hope to post a lot later on the ins, outs and ramifications of last night’s Labour disaster, but this morning if I may I’d like to make one key observation.

30 years in market research at the sharp end taught me one thing above all others: ignore what the respondents say, and watch what they do. In research, an audited record of consumer buying behaviour is worth a million focus groups because it is real. Upon that result, one can build norms, predict future likelihoods, learn solid marketing lessons….and grasp that the consumer is infinitely more (or less) discerning than you thought.

Research really gets into trouble when it tries to predict based on attitudes and opinions expressed. The simple truth about spending a career interviewing the public is that, for the practitioner, the skill is in working out when people are lying to you. In a politically charged atmosphere, there will be endless social emotions producing ‘white lies’.

This is why, in a nutshell, the BBC exit poll based on verified votes cast was right last night – and almost everyone else (including me) was wrong.

The other dimension of this issue is the extreme care required in learning as much as one can about WHO the undecideds are, WHY they claim not to have made up their minds, and WHAT they voted in previous elections. My experience suggests that in a majority of cases, such people are reacting in one of two ways: first, “mind your own business”; and second, voting in a way that they think perhaps “reflects badly” on them….for example, Camerlot Conservative.

At the outset of the 2015 election, some polls were showing 25% ‘not sure’ how they’d vote; even on the eve of the real vote itself, 14% were still undecided. That is easily enough to turn a dead-heat into a majority…however slim. And although most Sloggers thought I was joking about inebriation, with the polls staying open until 10 pm these days, probably 1 in 10 voters after 6 pm were anything from fuzzy round the edges to legless. This is the point at which they vote for the Silly Party candidate Mr Biscuit-Barrel….and, I should add, on a more selfishly right-brain way. All this tends to understate the real Conservative vote.

And let’s not forget, 34% once again abstained, couldn’t be bothered to leave the sofa, or just felt alienated enough not to vote. (Like me)

More later hopefully.

Recently at The Slog: Why the disenfranchised deserve a say