Part 1: On the Uses and Abuses of Reason
SOMETHING CLEARLY WENT WRONG with food in both Britain and America between the 1950s and the 1980s: it came to be considered to be more about convenience than pleasure. It seems astonishing now, but the predictions of the future I read as a child assumed that meals would be replaced by parcels of nutrients consumed in handy tablet form â it was for some reason thought that the purpose of food was to provide the necessary minerals, vitamins, protein and energy, and that the job of the food industry was to supply them in as efficient a form possible.
Some forward-thinking people had defined foodâs function narrowly, in order to create a rational model of what the food industry should do.* In this focus on scale and efficiency, people lost sight of what food is for; while it is, of course, a form of nourishment, it also serves a host of other ends. The proponents of delivering food in pill form had lost sight of the fact that it is enjoyable to eat and a necessary prop at social occasions.* Even if such pills could be produced, it is perfectly plausible that people who ate only such food would be utterly miserable.
In many ways it is the very inefficiency of premium foods that gives them their emotional value. The sourdough bread beloved of hipsters is insanely slow and inefficient to produce. Likewise it is absurd for the French to have so many local varieties of cheese, and yet this variety and scarcity seems to add to our pleasure. Contrast it with the US cheese industry thirty years ago â which was fabulously efficient and centred on a small number of states. In the 1990s there seemed to be only two varieties of cheese, a yellow one and an orange one, and neither was much good. Similarly, before the recent revolution in craft beer, the range and quality of American beer was dismal;* however, since American brewing has become magnificently diverse and inefficient, the US has gone from being the worst country for a beer drinker to visit, to the best.*
Food has become remarkably inefficient, and the pill-promoting futurists of the 1960s would be astonished to see how wrong they were. People spend hours preparing it, eating it and watching television programmes about it. People cherish local ingredients, and willingly pay a premium for foods produced without chemical fertilisers. By contrast, when we made the food industry logical, we lost sight of the reasons we value food at all.
Using this as a metaphor, I would like to see the improvement we have enjoyed in food over the last three decades applied to other fields. It is only when we abandon a narrow logic and embrace an appreciation of psycho-logical value, that we can truly improve things. Once we are honest about the existence of unconscious motivations, we can broaden our possible solutions. It will free us to open up previously untried spaces for experimentation in resolving practical problems if we are able to discover what people really, really want,* rather than a) what they say they want or b) what we think they should want.
1.1: The Broken Binoculars
For the last fifty years or so, most issues involving human behaviour or decision-making have been solved by looking through what I call âregulation-issue binocularsâ. These have two lenses â market research and economic theory â that together are supposed to provide a complete view of human motivation. Thereâs only one problem: the binoculars are broken. Both the lenses are pretty badly cracked, and they distort our view of every issue.
The first lens is market research or, to give it a simpler name, asking people. However, the problem with it is that, if we remember David Ogilvyâs words: âThe trouble with market research is that people donât think what they feel, they donât say what they think, and they donât do what they say.â People simply do not have introspective access to their motivations. The second lens is standard economic theory, which doesnât ask people what they do and doesnât even observe what they do. Instead it assumes a narrow and overly ârationalisticâ view of human motivation, by focusing on a theoretical, one-dimensional conception of what it believes humans are trying to do. Again, behavioural economics has shown that it provides an incomplete and sometimes misleading view of human behaviour â neither the business nor the policy worlds have paid sufficient attention to the failings of economics and research. Why might this be?
Generally, it is safe for anyone making business or policy decisions to act as though everything seen through these binoculars is accurate â not least because everyone else they work with â and everyone who might hire, promote or fire them â sees the world through the same binoculars.
âThe economic model told me to do itâ is the twenty-first-century equivalent of âI was only following orders,â an attempt to avoid blame by denying the responsibility for oneâs actions. Sometimes the old binoculars work well, of course: quite often people can accurately describe their motivations, and a large part of human behaviour is perfectly consistent with economic theory. Logic and psycho-logic do overlap frequently, as you would expect.
However, we still need a new set of lenses; as I explained at the beginning of the book, stubborn problems are probably stubborn because they are logic-proof. In other cases, the old binoculars provide a view that is so distorted, a field of view so narrow, that they blind us to far simpler creative solutions. The broken binoculars assume that the way to improve travel is to make it faster, that the way to improve food is to make it cheaper and that the way to encourage environmentally friendly behaviour is to convert people into passionate environmentalists. All these ideas are sometimes true â but not always.
Any new binocular lenses provided by sciences such as behavioural economics and evolutionary psychology will not be flawless, but they can at least provide us with a wider field of view. All progress involves guesswork, but it helps to start with a wide range of guesses. The following is a simple example of how a new lens can allow you to see (and solve) problems from a more psycho-logical perspective.
One of our clients at Ogilvy Change is a large energy provider who arranges appointments with customers for engineers to repair or service their central heating boilers. The appointments are scheduled either for the morning or for the afternoon â it is difficult to be any more precise than this, since it is hard to predict how long each visit may take. Customers complain about this, their most common refrain being, âI had to take the whole day off work.â What these customers say they want is a one-hour appointment window. However, if you were to take their demands literally and attempt such a level of precision, it would cost a fortune and there would be a risk of disappointment whenever circumstances prevented an engineer from delivering on the promise. The more astute of you may also have noticed that the one-hour appointment window would not necessarily solve the problem of âhaving to take a day off workâ â if your appointment was between 1pm and 2pm, for instance, unless you worked a short distance from home, youâd still have to take a day off work to be available.
Our first recommendation to the client was to listen to what consumers said, but to interpret it laterally rather than literally. People clearly found something about the length of the appointment window annoying, but maybe it was the degree of uncertainty involved in waiting for the engineer to show up rather than the length of the appointment window. Anyone who has waited at home for five hours for an engineer knows that itâs a form of mental torture, a little like being under house arrest; you canât have a bath or pop out for a pint of milk, because you fear that the second you do, the engineer will turn up. So you spend half the day on tenterhooks, afraid that your engineer might not show up at all. How different might the experience feel if the engineer agreed to text you half an hour before showing up at your door? Suddenly youâd be free to get on with your day almost as if it were a day off, with your only obligation being to keep an eye on your phone.* This is one of the solutions we propose to test. Is it as good as offering one-hour appointments? Not quite, but it might offer 90 per cent of the emotional and perceptual improvements, at 1 per cent of the cost. The old binoculars would not have revealed this because they would have taken customer complaints literally.
My colleague Christopher Graves, who founded the Ogilvy Center for Behavioral Change in New York, calls this approach âasking the real whyâ. People may be accurate commentators on their emotional state, but the causes of that emotional state (in this case, uncertainty) are often a complete mystery to them. If the experiment works, and early indications are positive, we have performed a form of alchemy, using psycho-logic to conjure up value from nowhere. Experimentation is the only reliable way of testing, so we measure the effect of engineersâ texts on customer satisfaction against a control group who receive no such early warning.
Another method is to perform what is called a thought experiment. For instance, ask yourself which message on a flight departure board would distress you more:
BA 786 â Frankfurt â DELAYED
or
BA 786 â Frankfurt â DELAYED 70 minutes.
The second message is a bit of a pain â but at least you are in control of the situation. You may need to make a few apologetic telephone calls, or go to a lounge and get your laptop out, but you can get on with re-planning your day. The first message, however, is a form of mental torture. You know there is bad news, but you do not have sufficient information to respond to it. Is it a 10-minute delay or a 90-minute delay? You might also worry that âdelayedâ is merely a precursor to âcancelledâ. That loss of power and control can create far stronger feelings of annoyance than the loss of punctuality.*
Unfortunately we are unable to distinguish between these two emotions: you donât say, âI am unhappy because inadequate information has left me powerlessâ; you say, âIâm angry because my bloody planeâs late.â In such cases, neither lens of the binoculars will present you with a solution. Airline passengers wonât want me to say this, but itâs true: if, as an airline, you have a choice between delaying a flight by an hour or spending ÂŁ5,000 to leave on time, your decision should be influenced by the quality of passenger information you can provide. I would also say that, from a psycho-logical point of view, metrics which target the punctuality of an airline without factoring in the quality of their information may be encouraging them to optimise the wrong thing.* (Remember also that perhaps twenty passengers on any flight might be delighted to receive a text message telling them that their flight is delayed â namely the people running late.)*
This might all sound like rather a trivial use of behavioural science. But, as you will learn later, the same techniques which can solve minor problems can also be deployed to solve much larger ones. For instance, the technique which might solve the problem of appointments for heating engineers may reduce peopleâs reluctance to save for their pension.* One of the reasons I believe there is genuine value to the study of behavioural science is that the same patterns recur: a solution which at a relatively trivial level helps encourage people to apply for credit cards might also be used to make people less reluctant to have medical tests.
More on this later . . .
1.2: I Know It Works in Practice, but Does It Work in Theory? On John Harrison, Semmelweis and the Electronic Cigarette
The approach that I am proposing will help you generate new and interesting ideas that are worthy of experimental testing, but do not expect them to be immediately popular or easy to sell. If you would like an easy life, never come up with a solution to a problem that is drawn from a field of expertise other than that from which it is assumed the solution will arise. A few years ago, my colleagues produced an extraordinary intervention to reduce crime. They hypothesised that the presence of the metal shutters that shops in crime-ridden areas covered their windows with at night may in fact increase the incidence of crime, since they implicitly communicated that this was a lawless area.
One of my colleagues, the brilliant Tara Austin, had seen research that suggested that âDisney facesâ â large-eyed human faces with the proportions of young children â seemed to have a calming effect. Combining the two ideas, she created an experiment where shop shutters were painted with the faces of babies and toddlers by a local graffiti artist collective.
By all measures, this seemed to reduce crime significantly; moreover, it did so at a tiny cost, and certainly by less than the cost of direct policing. Several other local authorities have since repeated the approach, though take-up is low â it is much easier to argue for larger policing budgets or for the installation of CCTV, than to approach a problem psycho-logically.
Copyright © Benoit Grogan-Avignon, with permission of Shutter Media
Contrast this effect with the impression given by plain steel shutters.
In a sensible world, the only thing that would matter would be solving a problem by whatever means work best, but problem-solving is a strangely status-conscious job: there are high-status approaches and low-status approaches. Even Steve Jobs encountered the disdain of the nerdier elements of the software industry â âWhat does Steve do exactly? He canât even code,â an employee once snootily observed.
But compared to an eighteenth-century counterpart, Jobs had it easy. In the mid-eighteenth century, a largely self-taught clockmaker called John Harrison heard that the UK Government had pledged ÂŁ20,000 â several million pounds in todayâs currency â as a prize for anyone who could establish longitude to within half a degree* after a journey from England to the West Indies, and was determined to find a solution. This was a life and death matter â a navigational disaster by British naval ships off the Isles of Scilly in 1707 had left several thousand sailors dead. To judge proposed solutions, the crown established a Board of Longitude, consisting of the Astronomer Royal, admirals and mathematics professors, the Speaker of the House of Commons and ten Members of Parliament.
Youâll notice that there were no clockmakers on the committee â the prize was clearly offered under the assumption that the solution would be an astronomical one, featuring celestial measurement and advanced calculation. In the end, H...