Marketing and the science of ‘knowing what economists are wrong about’

Rory Sutherland, the British advertising veteran, rails against a business culture that preaches rational, data driven decision-making over all else

It would be a mistake to take Rory Sutherland at face value. Walking through the lobby of a self-consciously posh London hotel, the British advertising industry veteran makes for an unlikely rebel, dressed in primary colours and speaking like Stephen Fry on speed.

But this self-styled “fat conservative white man” (his description) is a radical alright, railing against a business culture that preaches rational, data driven decision-making over all else.

Sutherland (54) has spent his career in the world of London ad agencies. When he started, the job demanded a constant supply of ideas, the bigger the better, around which long-running and expensive campaigns could be constructed. Television commercials were all but unavoidable, so the work of the copywriter became a part of mainstream culture. Vorsprung durch Technik; Snap, Crackle and Pop; Beanz Meanz Heinz; Finger Lickin' Good: Brand names were made famous and the Mad Men were rewarded accordingly.

The accountants hated it because the impact of TV advertising was notoriously hard to measure – "I know half is working, I just don't know which half" was the industry motto – but the spreadsheet monkeys had little choice than to take the seers of Madison Avenue and Soho at their word.


Then the internet happened, and our attention was dispersed across a whole new set of screens. This has allowed Google and Facebook to dominate the supply of eyeballs to those pushing breakfast cereal and German cars.

Silicon Valley’s sales pitch was an accountant’s wet dream, offering the promise of a direct communication channel between seller and buyer. The uncertainty of TV targeting was replaced by a new mantra of “measurable customer engagement”.

The creatives are under threat from clients seeking certainty: Don’t just come to me with an idea, say the bosses of today’s multinationals. Bring me data. Evidence.

Enter Rory Sutherland, waving a book with a golden cover.

“When you demand logic, you pay a hidden price: you destroy magic,” he writes in Alchemy: The Surprising Power of Ideas That Don’t Make Sense.

“We’ve developed models for looking at the world which aren’t necessarily bad but have been too universally used and believed,” he says. “Just as the map is not the territory, the tube map is a good way of thinking of the Underground system, but it’s a very bad map of London. Economics is to the real world of business what the Tube map is as a guide for getting around London. It’s sometimes useful and sometimes woefully misleading. When everyone’s using the same map it becomes more important for the things it leaves out as the things it contains.”

Behavioural economics

Economists in particular don’t come out well from an hour in Sutherland’s company.

“My definition of marketing is simply the science of knowing what economists are wrong about,” he says. “The human mind does not run on logic any more than a horse runs on petrol”.

This sort of thing is just the start of a hugely entertaining romp through ideas big and small.

“I think large numbers of people in business don’t really believe in economics. I don’t. I’m not even sure that in its current state it should exist at all. But the truth of the matter is that it’s much easier to go around thinking that everybody else believes it. It’s the safest position to share the opinions of everybody else.”

Much of the intellectual leg work within Alchemy has been done elsewhere, by people such as Daniel Kahnemann, Amos Tversky, Nassim Nicholas Taleb, Cass Sunstein and the other rock stars of behavioural economics.

Sutherland takes these insights and applies them to consumer behaviour, carrying out social experiments under the guise of well-funded marketing research. He has created a behavioural economics unit within Ogilvy and runs the annual Nudgestock event.

He says the sheer number of choices afforded today's shopper makes consumer capitalism "the Galapagos Islands for understanding human motivation".

“The rise of behavioural economics is incredibly necessary and arguably about 80 years overdue. The Austrian school spotted there was something wrong very early on. They realised that the mathematisation of economics was creating a model of rational economic man that tried to improve the world for a species that doesn’t exist.”

Beautiful theory, wrong species said Edward Wilson, the world's foremost expert on ants when he listened to someone describing Marxism.

“If we behaved in the way economists think we do, we wouldn’t have been able to evolve as a species. If anyone behaving in a way deemed economically rational would have been clubbed to death within a week. Basic things like the failure to reciprocate or the failure to share, which on the face of it might seem irrational and counter to our own success, are part of what makes a human society function.”


From Marx and Schumpeter, we're on to the plight of Danish opera singers, some of whom did very well until the invention of the gramophone. "There was a decent living to be made as the fifth best operatic tenor in Denmark. But suddenly their world changed when music lovers could buy records. If you're buying an album why wouldn't you want Caruso on it? Suddenly there were winner takes all effects, a phenomena that dominates today's internet business.

“We have to be really careful because all algorithms contain within them the assumptions of their creator about what’s important,” says Sutherland. “In this way, algorithms reveal the biases and prejudices of their creator. And they contain the wrong assumption that there is a single optimal answer regardless of who its dealing with.”

He cites Google Maps as an example. "It is designed by Californians, who have a whole set of assumptions about transport that a British or Irish person wouldn't make. One of them is that, broadly speaking, the only reason you take public transport is because you don't have a car. In California, this is a not perfect but reasonably robust assumption. But it doesn't understand that you might drive to a railway station and catch a train. For most people who live outside a major city, the best way of getting in to central London or central Dublin is to drive to a station and go from there. From a Californian perspective that would be completely mad. The other thing is that Google doesn't factor in journey time variance. If you've got a 9am meeting, what matters is the time of the worst-case scenario, not the average journey time. If my average journey time is half an hour, but worst case can be an hour and a half, then average is pretty useless."


Running through the conversation is a plea to recognise, and celebrate, that we’re not robots, and to acknowledge “the gulf between our unconscious emotional motivations and our post-rationalisations”.

“Measurebation is a phrase that emerged in the digital camera community which described people who were obsessed with the technical features of their cameras but who still took crap photos. They burbled on about the anti-alias ratios whilst churning out appallingly framed shots of people with their heads cut off. The theory said one thing, the reality was very different. This is what has taken over in business. The principal objective motivation is the avoidance of blame. The more you can cleave to numerical models – the less you are forced to use subjective judgement – the more you are likely to avoid being blamed if things turn out bad.”

In Sutherland’s hands, even working for The Irish Times becomes a case study in the psychology of decision making.

“Compare working for a reputable employer like The Irish Times to some dodgy tabloid startup. You might get offered a bigger salary, but the risk of it going under is far higher, so the downside variance is far more. Any decision where the outcome is unclear, you have to consider two variables: What is the outcome going to be like on average; and what’s the downside and upside variance? If the dodgy tabloid gives you chance of million quid in shares, then that’s different – bigger upside variance.”

This is why we buy brand names, he says, not because we love the product but because we think we can live with the downside. "If you buy a Samsung television, somewhere deep in your brain is making a prediction that the TV will be somewhere between okay and very good. If you buy an unknown brand, you're thinking this telly is going to be somewhere between OK and a crock of shit."