freshidea - Fotolia

Why users don't always know what they need

The economic concept of revealed preference can often tell IT managers more about what their users want than users themselves

Economists have a very useful little insistence, one they’re very fond of, which says: “don’t, whatever you do, believe what people tell you; watch what they do instead”. Rather than just expressing such cynicism, they dress it up in their beloved jargon – this is known as revealed preferences, rather than expressed.

It’s also something of great explanatory power. It might well be true that locals say they would prefer a vibrant High Street but the supermarket is probably onto a good bet with the idea they’ll flock to the out of town store for the lower prices. This has historically proven to be the case.

So it is with much of the worrying over the tech scene. To take the case of Uber, for example, there is much agonising over its “bro” culture, the inherent sexism of hyper-aggressive males getting on with what they want to do in the pursuit of pelf and lucre.

Yet this worrying seems to be coming from those who generally agonise over such things. The general public don’t seem to give Rhett Butler’s damn – we’re still flocking to the company’s product, those cheap and clean cabs, however kind they are to the female engineers – or not, as the case may be.

It’s true that former CEO and founder Travis Kalanick has gone, but that appears to be rather more to do with investors’ fears that he might not start making them money than anything else.

Hand over your data

A similar point can be made about the use of data by Google and Facebook. Those who don’t like US tech giants – much of the European elite, for example – seem to think everyone should be very concerned. Some indeed are, and battle with adblockers, refusing cookies, and so on.

The vast majority are delighted to hand over information, of no value to themselves, in return for the free services they like. One interesting proof is that, from memory, Bebo and MySpace were far less intrusive about their collection of data but we all left them as howling wastelands to join Facebook.

AltaVista collected less data than Google, and, well, the observation of what people actually do shows they just don’t care very much about this, and are even delighted with the deal on offer whatever their betters tell them they ought to think. All of which makes a bit of a mockery over the EU’s insistences on data protection and don’t you go and process it in the US. No one cares, see?

Another insistence says Twitter has a higher penetration among African Americans, and thus there should be a more diverse content team working at Twitter. The very fact of the greater penetration already existing seems to militate against the idea that people care about who rather than what – although greater diversity remains a valid goal.

A corporate boo-boo

Look at what happened years ago to New Coke. For the young shavers among us, the Coca-Cola Company thought that, after a century, they might try changing the recipe. Broad taste tests were carried out with huge market research efforts, all telling the drinks maker people would flock to buying a slightly sweeter (a bit more like Pepsi) version. So New Coke was rolled out nationwide, at which point sales slumped and the company looked to have made one of the biggest boo-boos in corporate history.

It seems those who liked a sweeter version, more like Pepsi, bought and drank Pepsi anyway. Thus the old recipe came roaring back as Classic Coke.

Read more about IT and economics

The point of this being that again it wasn’t any use listening to what people were saying. What people do can be and often is entirely different from what they tell the person with a clipboard. Or even what they tell their partners, friends and children. Who would say they prefer enriching the Walton family by shopping at Asda rather than that lovely nice butcher, with their handmade sausages, whose kids we see at school? And who doesn’t slip over to Asda all the same?

Much the same applies to politics – pre-Brexit focus groups didn’t quite manage to find that mulish streak of “sod the foreigners” which did rather drive the referendum result.

Market economy

We can even extend this out to how we must operate the economy. If asking people what they want doesn’t work, it’s rather difficult for us to plan anything. Many would indeed say they’d be perfectly happy living in a flat rather than a house, for example. Indeed, many did when they were asked. But as soon as the blocks went up, we found out that, by looking at what people tried to buy or rent, the British are very attached to the private house with a little piece of garden.

If it is revealed which preferences actually matter, we have to have a market economy, for that’s the only manner in which people do try things out and see who likes them.

Note that this applies to corporate planning as much as government. New Coke was a planned operation, a bureaucracy swinging into action, just as much as Ronan Point, the East London tower block that partly collapsed in 1968, two months after opening.

One way of thinking of this, a useful one at times, is that the market is the experimentation machine. Tastes change; what it is technologically possible to do changes; so we need some method of rolling through all the possibilities to see what it is that people actually want. To do so, we have to present them with the opportunity to have it and see what they do.

Another, and non-conflicting, manner is to agree planning is going to have to happen. Producers do of course plan what they’re going to produce. The market being the sea upon which the various plans sail, most of them to sink without regret, and a few survive. But given that our entire aim is to get to people as much as the people want to have what they desire, they must in fact have the choices with which they can reveal their preferences.

Read more on CW500 and IT leadership skills