The benefits of ignoring what our customers say

If we want to understand our customers, we can’t go wrong with asking their opinions. But relying solely on self-reported wants can be a costly disaster, as Burger King and others discovered when their customers told them they wanted healthier options.

We’re a highly inferential species. The way a question is worded and presented is information in itself, and can push us to give the answers we think the questioner wants. But we should also try and support any customer surveys with data on actual customer behaviours. Because they don’t always align.

It’s not always easy to articulate what we want

In the early 90s, the UK magazine sector for young men was a bit moribund. When surveyed, young men would list their interests – video games, sport, cars, music – and specialist magazines catered to them. But sales were generally low and dropping.

Received wisdom had it that young men just didn’t read magazines much anymore. Then some bright spark had the idea to look at what young men weren’t telling the market researchers.

That they liked looking at pictures of women without many clothes on, but felt ashamed of buying pornographic magazines.

This insight was translated by magazines like FHM and Loaded into: magazines for normal guys who like sport, music, video games and looking at pictures of women without many clothes on.

The emphasis was very much on ‘normal guys’.

This insight – which I’m sure would never have been discovered in a generic survey of interests – was that young men liked looking at attractive women, but not so much that they were willing to be perceived as weirdos by the people working in their local newsagent.

FHM and Loaded, by packaging attractive women as just one of young men’s interests, and doing so with (now quite dated and misogynistic) humour, created an entire category that sold millions of premium-priced magazines.

The fact that the internet has killed the category stone dead proves the power of the insight. Young men today can bypass the newsagent entirely when they want to see women wearing fewer clothes, so no one buys those magazines any more.

We didn’t evolve to articulate what we want

There’s no one reason why our potential customers can’t articulate what they want. They may not know themselves.

There’s a famous quote attributed to Henry Ford (who had a very poor opinion of customer surveys) that goes: “If I asked my customers what they wanted, they’d have told me faster horses.”

There’s also an argument that suggests that evolution may have been more strongly driven by spontaneity than self-knowledge.

Many animals that need to escape from predators seem to rely on movement that appears completely spontaneous. And it’s not hard to see how, were an animal to plan each escape-move, a wily predator might learn telling body movements (like muscle tension) and be more likely to catch the animal rather than less.

It’s certainly true that we’re all influenced by competing drives. The wants of the pre-frontal cortex, which allows us to imagine our future and plan for it are often overridden by the ‘lizard brain’. The brain stem, cerebellum and basal ganglia (the structures that comprise the lizard brain) are also common to far simpler animals.

The lizard brain has helped keep us alive in more precarious times. Historically (even up until not much more than 100 years ago) sources of sugar were rare. When the lizard brain encounters sugar, it doesn’t want to think. It wants to consume it.

Useful behaviour for our ancestors on finding a honeycomb. Not so useful when our long-term goal is to lose 10 kg.

There is actually an argument that language evolved to help us (as social animals) post-rationalise and justify the crazy things we’ve done after the event.

No one wants to admit to being driven by primitive drives

Legendary adman David Ogilvy said: “The trouble with market research is that people don’t think what they feel, they don’t say what they think and they don’t do what they say.”

But why? All living creatures with any degree of agency are driven by the Four Fs (fighting, fleeing, feeding, and reproducing). Some participants might feel that base drives are just something polite people don’t talk about.

But it might be more than that. The lizard brain is all about ‘now’. It doesn’t have time to help us understand why it wants what it wants. And if we don’t know, how can we articulate it?

By way of example, what first attracted you to your partner? There’s a legendary study from the 90s that suggests that Major Histone Compatibility (MHC) could have been a significant factor. Don’t recall noting how genomically-compatible you and your partner were when your eyes first met at the office Christmas party?

Unsurprisingly, it was most likely subconscious.

Claus Wedekind, who led the study, hypothesised that we might be able to detect compatible genes in a potential mate through odour. He got a cohort of young men at a university to sleep in the same t-shirts for a few nights. Then he took their t-shirts and gave them to a cohort of young women. The women consistently preferred the odour of the t-shirts worn by men whose immune genes were different to their own. This implies that we might subconsciously be able to detect a compatible partner for having children.

There’s a ton of study in this area now. It suggests that we may have our reasons for preferring one partner over another. But we wouldn’t be able to explain why.

The online challenge

Where our products are online, it’s both a blessing and a curse. On the one hand, we can gather swathes of comparable data without much effort. On the other hand, we can’t often observe our customers ‘in the wild’ and gain insights into their behaviour.

For example, if two users take five minutes to go from sign-up to their first game of black jack, they’ll look very similar in the data. But one of them might have been totally confused as to how to proceed. The other might have found our interface so dull she got distracted by Match of the Day.

This likely makes working to hypotheses based on evolutionary psychology more important than ever. It’s difficult to make sense of data. It’s hard to rely entirely on customer feedback. But using our evolutionary history as a starting point can help to put these into some perspective.