Rory Sutherland on How Can We Use Tech to Hack the Human Consciousness?

Rory is a world class talker and takes some shutting up. In fact, we know that which is why we scheduled this talk to be flexible. That’s why it went on for 90 minutes and no one left the room…

Rory is Vice Chairman at O&M Group UK, an attractively vague job title which has allowed him to co-found Ogilvychange, a behavioral science practice. You should watch this talk if you want to change your perspective on product and problems.

Rory co-heads a team of psychology graduates who look for “butterfly effects” in consumer behavior – these are the very small contextual changes which can have enormous effects on the decisions people make – for instance tripling the sales rate of a call centre by adding just a few sentences to the script. Put another way, lots of agencies will talk about “bought, owned and earned” media: we also look for “invented media” and “discovered media”: seeking out those unexpected (and inexpensive) nudges that transform the way that people think and act.

The unconscious plays a much great part in decision making than we would like to admit. Most exciting technological innovations succeed because they are rationally better in some dimension but unbeknown to the people who have instigated them they have an almost magical power to settle the mind.

Find Rory’s talk video, slides, notes, transcript, and more from Rory below.

Find out more about BoS

Get details about our next conference, subscribe to our newsletter, and watch more of the great BoS Talks you hear so much about.



Learn how great SaaS & software companies are run

We produce exceptional conferences & content that will help you build better products & companies.

Join our friendly list for event updates, ideas & inspiration.

Unsubscribe any time. We will never sell your email address. It is yours.


  • Maybe you could improve business by closing all the business schools. It’s not that they are useless, they just all teach the same stuff.
  • Economics is flawed, they develop theories and then when the theories don’t hold up, they blame the people.
  • How would you compete with Coke?
    • B-School answer: “No one can ever beat coke…” but if you did it would have to taste great, offer more, and cost less
    • Red bull has competed well!
    • Monkey brain thinks: if it tastes bad it has medicinal value because
  • Wagamama
    • “Food arrives in the order it is done, not when you expect it” – reframes the experience
    • Economics thinks efficiency is ultimate
  • Placebo choices
    • Still or sparking vs tap water
    • People like what they choose
      • Bullshit can help people
  • Five Guys…
    • They ask S, M, L… fill the cup up, then dump more in the bag.
    • Logic says “make the cup larger”
    • Accountant would say they are wasting money
    • Something feels good about getting more …
  • The human brain hates uncertainty…
  • Behavioral economics requires an understanding of evolutionary psychology
  • Human brains are optimized for making decisions in situations with a high degree of uncertainty, lacking trust and complete information.
    • Books: The Folly of Fools
    • The mind helps remove ambiguity.
  • Donald Hoffman’s evolutionary model
    • Your mind only cares about fitness
  • Herbet Simon – big figure in behavior science and AI
    • You can never solve a problem perfectly… you can only satisfice
  • Mathematically pure answers aren’t valuable if they are solving the wrong problem
  • McDonalds
    • Average and peak is low, variance is low
    • Good representative of the human instinctive form of satisficing.
  • What is the brain calibrated for?
    • Standard response: “To maximize”
      • Archery … there is no variance the rule system is standard. Aim for the bullseye. Archery does not make a great televised sport
      • Darts… more representative of reality. It has a much more variable scoring system. Most people aim for the perfect triple 20, but the better target is the southwestern target where you can’t get a 1 or 5.
    • Smoke Detector – Better to go off as a false positive from cooking than a false negative of not going off when it should.
  • We are calibrated to be disproportionately acutely attuned to recognizing faces of humans AND animals. There are many evolutionary advantages for this.
    • People therefore see faces in many places where they don’t exist
    • Cat and Cucumber-
  • We follow paths because we assume others have gone that way, which likely means it hasn’t resulted in catastrophe
  • A lot of human decision making is also about not looking stupid if you fail… it leads people to go the conservative route.
    • “No one ever got fired for buying IBM”
    • Soccer FGs kick left or right, when likely straight wlil result in a goal because the goalie either jumps left or right.
  • Brand Preference
    • Not a very good guarantee that something is perfect BUT it is a guarantee it likely isn’t horrible. This is partly because a brand has more to lose by a bad product.
    • Monkey brain feels more comfortable buying a recognizable brand.
    • The difference between a good and very good television is irrelevant, the difference between an OK and bad television is massive.
    • “Upfront expense = costly proof of long term commitment”
      • Example = great expense associated with TV commercial
      • Example = engagement ring
  • AppleTV
    • When you advertise… it proves you are willing to put the scarce resources of money and attention behind something
  • A flower is a weed with a marketing budget
    • Biologists can understand more than economists and marketing because nature very instinctively represents decision making and risk
    • If you were given the same option presented by a priest or a man in his underpants… it impacts our perception instead and amount / willingness to pay because the guy in his underpants has no shame…
  • Satisficing results in higher happiness.
  • Emotions and heuristics can be inherited, reasons have to be taught.
  • Market research rarely focuses on talking with the decision maker
  • SCARF: David Rock’s Model
    • Status
    • Certainty
    • Autonomy
      • Matters for its own sake.
    • Relatedness
    • Fairness
      • Higher order primates that operate in groups also exhibit a value of fairness.
      • Monkey: Cucumber / Grape experiment
  • In business, it is considered a greater win to win through reason vs. emotion
    • Dyson example… watching the dirt accumulate in the container satisfies the monkey brain.
    • NOT SAYING he could have sold crap containers… had he made the vacuums opaque and vague, he would not have sold as much
  • “The worst thing British Airways could do is show DELAYED without an amount of time… it increases uncertainty and is very frustrating.
  • Uber … great mind hacks that increased the delight of the experience
    • Showing # time for pickup
    • Decreased uncertainty of watching the cab show up in real time AND you have information about the driver.
    • Paying with cash and having to deal with receipts is annoying
      • Better to separate payment from consumption.
  • Train example: putting WiFi on the trains increases the quality of the experience (marketing solution) vs.. increasing the speed by X % (engineering solution).
    • Compete with air travel by increasing the quality of experience.
  • Solving problems psychologically is viewed as cheating.
    • The placebo effect actually works!
  • Yield management practices… ??
  • Building in doubt can have an intended effect…
  • Evolution… make pretty good, non-disastrous decisions.
  • When information is presented in a crap order… people that designed it and customer don’t notice.
    • Flight travel example: asking about the flight class BEFORE knowing the price difference.
    • You can manipulate people without them realizing it…
      • Example: Wine is most profitable drink because it has no price ancho
        • Put wine glasses on the table increases purchase of win.
        • Provide a wine list instead of a drink list
  • If you want to increase diversity…When you are hiring one person, you go conservative. If you hire in groups, you are willing to take a risk.

Learn how great SaaS & software companies are run

We produce exceptional conferences & content that will help you build better products & companies.

Join our friendly list for event updates, ideas & inspiration.

Unsubscribe any time. We will never sell your email address. It is yours.


Rory Sutherland: Oh! Perfect! Thank you very much indeed!

Very simple title, the science of knowing what economists are wrong about.

There are some ideas in business which aren’t so much bad because they’re bad, they’re just bad because they’re universally believed. I had a conversation with someone from the London business school the other day who thought maybe you can improve business by closing down all the world’s business schools. This isn’t because what the world business schools teach is wrong, it’s that they all teach the same stuff and producing thousands of people who all think the same way is not a great way to produce a fantastically diverse and ecologically rich business climate!

And what I’d like to do, I consider consumer capitalism for all its many faults it’s also a massively well-founded social science experiment. And it’s really the Galapagos Islands of human psychology. And what I’d like to do is look at things that are very, very strange and I’m trying to explain them. If you’re an economist you will notice that economics is a very weird science because it cheats. What it does is it has a theory on how people ought to behave and when people deviate from that behaviour it doesn’t blame the theory but the people. You don’t have that luxury in physics, you can’t say well our experiment has failed completely but it’s due to the presence of irrational electrons. You’re not allowed to do that, but in economics you get to do that and in any way the people deviate from that theory, then you actually just blame people. I didn’t agree with that, I think we should at the weird things people do and first I will ask you if there’s some sort of deeper evolutionary sense behind them or at the very least try and work out what’s going on.

So I particularly like to make a study of things that are kind of bonkers and I will give you a most extreme example of this. If you go to a lot of rational people 15 years ago and say we want to compete with Coca Cola, it’s been there for pretty much 140 years, the world’s most successful cold non-alcoholic drink apart from water and we’d like some of that success. Every single person I think would come up with the same answer, it’s quite easy to compete with them, you just need to produce a drink that tastes nicer than coke and it needs to cost less than coke and it needs to come in a really big container so people get great value for money. And everybody in the board would applaud because it’s completely consistent with classical economic theory of how people behave. Meanwhile the most successful attempt to compete with Coca Cola in over 100 years is this, it costs a fortune, comes in a tiny can and it tastes disgusting. By the way, when I say it tastes disgusting it’s not a subjective opinion, they took the drink to a company which only researches carbonated beverages and the company reported back that this drink actually fared worse in research than anything else they have ever tested ever [laughter]. Usually people would say things like ‘it’s not really my kind of thing’ or ‘it’s more for kids’. In this case the respondent’s verdicts ranged from things like ‘absolutely repellent’ to ‘I wouldn’t drink this piss if you paid me to’. What’s going on? I don’t know, ok?

But I’m gonna leave you with a bit of a clue here I think which I will wrap up with if I have time at the end. I think there’s something deep in human instinct which believes that if something is to have medicinal or psychotropic powers, it has to taste weird, that largely explains the health food industry. Somehow the brain is comfortable when there’s a bit of a trade-off. I don’t really trust Neurofen Milklets cause they’re too tasty to be an effective analgesic. If you want something to be an effective pill, it’s got to be kind of repellent. Similarly, SE Johnson our clients discovered that you can in fact make a fly spray which kills flies and also smells like lavender so you can scent your house and kill insects at the same time. You can make this, but you can’t sell it cause the monkey brain doesn’t get this, it goes no, this isn’t right. Why am I going around scenting flies? Conventional fly sprays on the other hand fits beautifully with monkey logic, smells nasty to me so it must be a bastard if you’re a fly.

On the tiny things, those of you who are familiar with wagamama – everybody was there at some point? At your first visit or you pretend it’s they will ask you when you go in is this your first time to wagamama? If you say yes, they say what you need to understand is wagamama is based on authentic Japanese noodle bar and so the food arrives as soon as it’s ready fresh from the kitchen, rather than the order you expected to arrive in, ok? Now I would argue that rather bizarre sentence probably saved wagamama. The thing is this is entirely mental reframing, that sentence doesn’t change reality at all. It simply changes how we react to reality. If you didn’t utter that sentence, probably the majority of people who visited wagamama the first time if asked would have said place is total chaos, food arrived completely at random, I had a massive great bowl of noodles and just before my coffee arrived, the food showed up. People don’t know what they’re doing place is totally random, ok? Utter that sentence, everybody goes marvellous! This is just like a Japanese authentic noodle bar, the food comes straight from the kitchen! You’ve taken the reality and changed the way people react to it. You haven’t changed the reality at all.

Incidentally I gave a talk to someone in the audience who is Japanese and he claimed that it’s bullshit other than in real Japanese noodle bars, the food actually arrives in a pretty good order. Nonetheless, maybe it doesn’t matter. Maybe bullshit’s ok. Obviously you’d expect an advertising man to say that but I will give you a little example; talking to the public sector and the problem with it isn’t that it’s inefficient, that’s an economist explanation, they think that efficiency is the highest virtue for any new organisation. My contention is that most public sector organizations and most monopolies are bad because they don’t learn accidently to bullshit. My argument when I spoke to some people from the NHS was look, this is how you handle a phone call. Someone rings up and say I need an appointment for a knee operation, and you say we have an appointment on June 22nd in the morning. Ok? That’s how it works. So don’t do it like that! I said this is how you do it, you need to add some bullshit. So the phone call comes in, I need an appointment for a knee operation. Would you prefer the morning or the afternoon? If they say the morning you say that’s good because we have an available slot on the morning of the 22nd of June, ok? If they say afternoon, say you can have an afternoon appointment but it would mean waiting till late July whereas if you’d like a morning one, you could have one on the 22nd of June. Which would you prefer? And 99% of them will say the morning and everybody ends up with an appointment on the morning the 22nd of June, the difference is in the first case they feel like a supplicant whereas the second case they feel like a customer.

It’s a placebo choice basically. It cons you into thinking you have a choice where really you don’t have much choice at all. There’s lots of placebo choice – still or sparkling is that, it’s your restaurant saying don’t get any stingy ideas about ordering tap water. It’s much harder to order tap water where the choice is still or sparkling than would you like some water, right? Red or white is a brilliant way of stopping people asking for more interesting drinks like gin. If you think about it, most of your parties you offer them red or white, I’m pretty willing to content that even those parties where nearly everybody drinks red, if you’d only offer them red wine, most of those people would have left your party and say in the car that wine was pretty rank, wasn’t it? Because they chose it themselves and they like it more. If you choose seat 70A on an aircraft you’re much happier than when you’re allocated seat 70A. We like things we’ve chosen, my view is you make people like things without any environmental destruction but simply by the generation of bullshit, maybe that’s really environmentally friendly thing to do.

Other bits of bullshit I really like, when you go to 5 guys and they say – any fans here? Ok, they asked you what fries do you want, small, medium or large and they take a cup of the appropriate size and then they fill it to the top and take an additional scoop – you’ve noticed this and they put it on the bottom of your brown paper bag. Now, totally rational person would say make the cups bigger and an accountant would say stop giving away the free fries because they’re costing us money. This is much better than either of those 2 alternatives. Why? Because our monkey brain disproportionately likes people who give us things they don’t have to give us. They have to fill the cup up to the top, there’s no getting away with skimping on that one. What the hell you think you’re doing? Making the cups a bit too small and then giving people more fries on top creates more happiness than if you make the cups larger. Now I don’t entirely know the reasons for this, they are embedded in the brain’s hardware, it has to do with how we react with social interchange and they go back hundreds of thousands or in some cases millions of years. But if you could understand some of these things better, then you could work with people better.

I had a wonderful conversation over lunch about pensions. And the interesting question is this – I should try this on this audience. How many of you if I gave you 500 quid in cash and said you can keep that money so long as you pay it in your pension before you get home? How many of you would know how to do that? It’s interesting, we’ve got 5 – this is a marked improvement on 200 people working for a retail bank where the result was 0 and another audience of 200 where once person raised his hand but he worked for Golden and Sachs, so he wasn’t representative of the average man on the street. The interesting thing is in order to pay extra money from your pension, if you’ve got 500 quid and want to buy an iPad it’s quite easy. You go online and you buy one or you go to an Apple Store. If you want to transfer £500 from your current account to your deposit account, I assume you will find that quite easy as well. Ok, if you were to transfer 500 into your pension which is the most sensible thing to do with the windfall of 500 quid, it’s spectacularly difficult. You have to go home, dig into a filing cabinet, it would involve a cheque – anybody know where your cheque book is? Haven’t got a clue, haven’t written one for about a year. And it would involve posting something to somebody. This not only discourages pension savings because it’s difficult, it discourages it squared. Not only is it difficult but the instinctive brain tends to assume that if I was to supposed to do that, they would have made it easy. Well they can’t be supposed to top of my pension on a whim because if that were normal human behaviour then a mechanism would exist which made it easy to do.

Then another fact about the human brain is that it effectively very simply hates uncertainty, if you send that £500 by post off to your pension fund, how do you know the money had arrived? You wouldn’t get a text or confirmation at all, you’d have to wait 6 months until your next completely incomprehensible pension statement arrived and you’d have to remember the cheque. Finally, with that 500 pounds, you get a tax rebate which is actually quite lavish but this is paid in your pension in a way that is completely invisible. This costs the UK government something like £40 billion a year, rebates paid into pension contributions. I would argue that’s about 1/3 of the NHS and it’s the least effective £40 billion the UK government spends by some margin. In fact, pissing it up the wall wouldn’t be much worse to be honest.

Because giving people money very, very easily and invisibly has an absolutely miniscule incentive effect on behaviour. In fact it would be more effective if at the end of each year they said you paid in £5000 this year, this means you’re entitled to a £2000 rebate in your pension but in order to claim it you have to turn up to the government offices, have electrodes attached to your genitals and suffer extreme electric shocks for 15 minutes. Then you’d really remember it, wouldn’t you? And you’d go well 15 minutes actually £2000 isn’t bad, is it? I’m gonna double my pension contribution then I will really screw the bastards next time.

The simple fact is if you try to model human behaviour on the assumption that everything an economist says is true you will go wrong. Now you must be interested wildly in lots of anomalies which anybody who is a web designer or UX expert has all the time. How many of you know this case, the $300 million button? I think it’s BestBuy, is that right? They would say a large e-tailer but I think it was BestBuy. This was a case where you previously had a choice after you filled your shopping cart of sign-in or register. And they noticed a huge number of people just abandoned their shopping carts, they added a 3rd button which was called continue or checkout as guest effectively. But adding that 3rd option they sold $300 million more merchandise the following year, directly attributable to that small button.

Now what’s peculiar about that actually? It’s slightly weird when you think about it, is you could argue yes, but you didn’t have the trouble of registration. The odd thing was of the people who bought by checking out as a guest, when invited to register post purchase, 98% of them did. So it was a past dependency effect, it wasn’t really an effort effect. And there are tons and tons of these butterfly effects dotted all over human behaviour, the tiny little changes you make. Google made $200 million by changing the shade of blue on the links. They noticed that the shade of blue they used in Gmail advertised sponsored links was slightly different to the shade of blue they used on the search page and the home page. And they decided to test which shade of blue was better. Making it slightly purple made them another $200 million a year.

There are cases I’ve seen in the UK putting an automatic space bar where people type in their mobile phone number, made one e-tailer another million pounds a year. If you look at your mobile phone number, it’s true if you’re in the UK and not an American if you look at the mobile phone number it’s very difficult to tell whether it’s right or not cause it’s 11 digits in a row. If you split it up 5-3-3 that little more level of uncertainty disappears which makes the difference for a significant people on the margin between buying and not buying. But I will talk about path dependency a bit later.

What I think is really going on here is you can’t really understand human behaviour without at least understanding behaviour economics or the study of ways in which human behaviour differs from conventional narrow logic and you can’t really understand it without understanding evolutionary psychology.

Now if you think about economics, it’s assumes for the purposes for mathematical neatness that everybody makes a decision in an environment of perfect trust. And just to add to that, the ability to optimize in all decisions however complex so it’s also assuming a limitless processing power for the human brain as well. Now there are two facts about that, one of which is it doesn’t really – the real world is rarely like that, the second thing is that it any case, we didn’t evolve in a world of perfect information and perfect trust, so our brains are not optimized for such a world. We’re optimized to make decisions in situations where there is inadequate information, a high degree of uncertainty and the possibility of mistrust.

Now, just to prove this, there’s a very simple fact. Evolution in designing the brain doesn’t care at all about accuracy, it cares about fitness. If it can misrepresent reality to you, in such a way that it improves your chances of survival or reproduction that’s how your brain will present information to you. It’s not remotely interested in objectivity, this idea that evolution gave us eyes a bit like light meters and ears like recording equipment is not true at all. It gave us a perception of the world that was most representative of survival. Now one of those important things about that is it removes ambiguity without us knowing about it, it actually fills in loads of information because the most important information is information you can act on. That will aid your survival – information that makes you go I don’t know what to do is of very little use in an evolutionary setting. So it makes us unusually decisive and I can prove this with this what is an utterly fantastic experiment.

At any one moment, we are being bombarded by sensory information. Our brains do a remarkable job of making sense of it all. It seems easy enough to separate the sounds we hear from the sights we see. But there is one illusion that reveals this isn’t always the case. Have a look at this. What do you hear? But look what happens when we change the picture. And yet, the sound hasn’t changed. In every clip, you are only ever hearing ba, with a b. It’s an illusion known as the McGurk effect. Take another look! Concentrate first on the right of the screen. Now to the left of the screen. The illusion occurs because what you are seeing clashes with what you are hearing.

So this is our equivalent of error correction. The brain sees a fa, hears a ba, rather than present to you confusing information, it makes a punt. You have no conscious control of awareness of its happening. It simply goes it’s more likely that I have misheard or mis-seen so I will override that b with an f. And it does that all the time, this is the point of the evolutionary model from Donald Hoffman – that’s where your perceptual mechanism only cares about fitness, that is the model that always wins. And actually our brains are filling in information and we think we can separate what we hear from what we see, taste, smell or hear from what we know and in fact we can’t.

I mean genuinely wine tastes better if you pour it from a heavier bottle. I think they’re quite clever today having blue glasses, water is more refreshing when drunk out of a blue container, that’s a pure better mind. Cadbury’s were bombarded with complaints that they’d sweetened their chocolate where all they’ve done is change the shape of the blocks to make them rounder. Rounder things taste sweeter even though the ingredients haven’t changed at all.

Our perception of price is hugely relative, not absolute.

That’s another evolutionary mechanism, it’s much more important for us to detect contrast than it is for us to accurately observe absolutes. And that doesn’t just apply to brightness in the shade, it also applies to price. Who’s got an expresso machine here? Insanely expensive, I got one myself and love it like a child. If you had to buy the coffee in a jar like Nescafe for an equivalent for an equivalent amount of caffeine, the jar will cost you about 40 quid and you would look at it on the shelf and go that is bonkers. But the great thing is you don’t know what an individual Nescafe cup costs so you don’t really have a frame of reference when you put that to the machine. So when you do put the 29p pot into the machine, your frame of reference is not Nescafe, it’s Starbucks. It’s 29p, that would have cost me 2.99 at Starbucks! This machine is practically making me money!

Rolls Royce and Maserati stopped exhibiting their cars and car shows and started exhibiting them at yacht and aircraft shows. If you’ve been staring at Leer jets all afternoon, a 350000 car is an impulse buy, ok? Well we didn’t buy a place this year, so I will have a couple of those.

And one of the most fantastic things I have a friend who is an engineer who was driven insane by the fact that his car seemed to be a better car when he had it washed. And he was convinced because he’s an engineer, there has to be an explanation that somehow the actor rubbing your car and tautens the panels so there’s less vibration. It’s total brain bullshit but it’s a great way to get a valid car. That’s a really good bargain, cause you actually get about 50% of the pleasure of a new car at a tiny fraction of the cost. It isn’t just a cleaner car, it drives better, it’s smoother and quieter, it accelerates and corners better because your brain is fusing different sources of information. Marketers have known this for years which is why you never go to a car show room and see the cars covered in shit, ok? Would you like to test drive our vehicle which stinks of dog? Not a good idea, right?

Now this is the – Taleb. The point is the things we can solve rationally with a perfect single answer, i.e. science, those things are very few and real life imposes lots and lots of problems which aren’t like that. Your maths cheater was cheating you when he sent you those examples with 2 buses leave a bus station at noon. One travels due north at a constant 40 km/h, the other one travels due west at a constant 30 km/h. That’s the totally mistaken pretense that maths can be used to solve real life problems. It can but only in that fantasy world where buses travel at a constant speed and in a straight line and leave at the same time. Most real-world problems are much much messier than that and there isn’t a single optimally right answer.

Anybody knows who that is? Herbert Simon. Any fans? I think he’s a massive figure both in the decision science world and in the world of AI. He made the point that in most real world situations, you can never solve a problem perfectly, you can only do what he calls not maximizing but satisficing. You can do it in 2 ways, either solve an imperfect problem that’s kind of similar perfectly, or you can take the real problem and come up with a pretty good solution to it. There is no single right answer.

Let me just explain the two methods of what you might call satisficing.

This is a sat nav, ok? Which often provides you with a perfect answer to a slightly wrong question. I don’t obey my sat nav when I need to catch an airplane on the airport. I put it on and tell it to take me to the airport and then I quite often ignore what it says. Why? Because the sat nav is saying on the bounds of probability what is the fastest way of getting to the airport. I need to do something different, it’s don’t miss the plane, therefore optimizing, expected speed on its own doesn’t provide a perfect answer, you need to consider variants. I don’t go on the M25, the M25 is 10 minutes faster on average and 10 minutes faster 95% of the time but when the M25 clogs up, I’m stuck there for 2 hours with nowhere to go and I missed my plane. If I go on the A25 it’s always 15 minutes slower, but it’s never an hour slower. And so what the sat nav is doing and be very careful in business decision making of allowing the mathematical purity of your answer to allow everybody to believe that you have the right answer. You may have achieved that at mathematical purity at the expense of ignoring variables which in some cases are more important than the ones you’ve actually considered.

The alternative in satisficing – in many way McDonald’s is the best restaurant in the world. That’s a scandalous thing to say, but it is the best restaurant and certainly the most successful restaurant in the world. Not because it’s very good. The average is pretty low, the peak is pretty low but the variance is very low. Nobody has ever been ill eating a McDonald’s. Michelin star restaurant it’s fantastic 9/10 and the 10th you spend time on the bog, let’s be honest about that! There’s quite a high variance in high end food. This is what you might call the instinctive human form of satisficing. I want something that’s pretty good, it’s reasonable but which definitely isn’t terrible, it’s not gonna make me ill, I’m not gonna be ripped off and significantly disappointed and in the absence of negatives as opposed to the creation of positives, it’s not a bad restaurant.

Now, I think there’s a really important question here which is what is the brain calibrated for?

Ok? Now, the standard assumption is that the brain is calibrated to maximize and that assumes that real world problems are a bit like archery. Now archery has a completely concentric scoring system, you aim for the bullseye that scores you 10. If you miss that you get a 9, then an 8 and then a 7. There is no strategy ever other than aiming for the bullseye, it doesn’t matter if you’re drunk, the board is moving, there’s heavy fog, it’s extremely windy, you still aim for the bullseye because there is no better thing that you can do than that, it’s inarguable. Because of that I think archery doesn’t make a great spectator sport. It’s no surprised that televised archery has never really broken through to the mainstream.

Real world problems are much much more like darts, where the scoring is messier, it’s not concentric. Now interestingly, most people when they play darts aim for the triple 20 and that’s because they want to pretend they’re a professional darts player and if you’re Phil ‘The Power’ Taylor and there’s not much wind, no fog, the board isn’t moving, you’re relative sober by darts playing standards, you are right if you’re very good at darts to aim for the triple 20. Most people who are less good or in situations of some uncertainty like wind or a moving board best place to aim is actually for the SW quadrant, you might get a triple line or triple 16 which isn’t much worse than triple 20 and you won’t get a 1 or a 5.

Now I would argue that the human brain is to a great extent calibrated to aim for the SW quadrant of life. Not to attain perfection because often, aiming for perfection might come with quite a high cost. Not least of course you got to compete with all those other mad people who are aiming for perfection. The human brain is largely calibrated not to optimise but to actually satisfice, to go for a pretty good solution which is not awful.

Now, if you look at the question of calibration your smoke detector is calibrated, it goes off when you make toast and you have a fry up. That doesn’t make it a bad smoke detector. You’ve got to have a degree of paranoia in a smoke detector because a false positive is much less severe in its consequences than a false negative, ok? If your smoke detector goes off when you’re making toast they are annoying but it’s a lot better to have a smoke detector that does that from time to time than to have a smoke detector which is a stoner or hippie smoke detector and goes like what’s that smell? Kind of weird, bit like smoke. I will just give it the benefit of the doubt. You don’t want your smoke detector to do that.

So a large part of the brain is just calibrated to avoid catastrophe much more acutely than it is calibrated to seek to attain perfection. If you look at the question of calibration, I put these in because I thought they’d interest you. We’re calibrated as human beings for very good Darwinian reasons, to be disproportionately acutely attuned to recognising faces, not only human faces but animal faces. And there are all sorts of evolutionary advantages to detecting faces well. You can recognise predators hiding in a bush, you can probably infer people’s emotional state from their expression, you can spot people or animals disproportionately well when the background is a lot of crowded foliage. It’s really useful to have our brains calibrated to detect faces. And like a smoke detector, they are calibrated to such an extent they will sometimes detect faces or inanimate objects where no such thing exists. It’s called pareidolia. This is where someone sees a coat hook – drunk octopus wants to fight [laughter]. You can’t help now but see that as a kind of entity, that’s pareidolia that bias, ok? Similarly when an architect decided to put rounded windows in a church in Florida he thought it would be known as the church with the lovely rounded windows. Is it known as the chicken church! [laughter] It’s impossible to look at that – and we also have other wonderful effects like a drunk washing machine for example.

Now what’s interesting which I thought my be interesting for you software folk is when you design facial recognition software, there is no alternative but for it to be calibrated occasionally to bring out false positives. If you’re doing facial recognition software that never comes up with a false positive, it’s really crap at recognising faces! You have to have that bias there and if you had software which never saw a face when there wasn’t one, it would recognise only human faces when they would look face on, perfectly complete and with no vague anomalies like spectacles or beards. You’ve got to give the calibration a bit of skew.

Now those of you with kids will know about faceswap software on the iPhones. Is that right? You’ve all been the victim of faceswapping. The facial swapping software, no less than the human brain, suffers from pareidolia. Now if you just pause and look at that for a second. Notice it doesn’t recognize that guy as a face because part of his eye is missing so it’s desperately casting around for something to swap around with a baby’s face and it finds it on the front of the cooker [laughter]. Either that or it’s a perfectly conventional photograph and there’s a family that needs some help.

But this by the way is how cats are calibrated like smoke detectors. If you can do this with your cat at home, it’s cruel but highly amusing – if you take a cucumber and you sneak up behind the cat outside and put it down on the ground outside its field of view, when the cat detects it, it doesn’t do a cost-benefit analysis. Right? What it does is it goes it’s green and got a pointing nose. Either it’s a snake or it’s crouching to pounce, I’m not gonna actually treat this in an even handed manner. I am gonna act like a smoke detector and assume the worst. And this is indeed what the cat does.

Now, evolutionary biologists explain this to be perfectly, this calibration to the human brain which is we’re much more acutely tuned to avoiding disaster than achieving perfection. The odd thing is the rational part of the brain doesn’t realise this, we don’t think the way we think we think because the bit of our brain which contributes consciousness is a very small and actually slightly misleading part of the brain. It does all the talking, it doesn’t do much of the deciding, more on that to follow.

But what is interesting, once you understand we’re trying to avoid terrible rather to achieve perfect, a lot of instinctive human behaviour stops being irrational and starts making sense. Habit, I’ve done it loads of time before. It’s always been ok, therefore I will do it again. Also the fact that it becomes habitual in life – you will notice this when you’re older when my kids say where do you want to go on holiday my answer is somewhere we’ve been before, I know where all the shops are. This is what happens when you’re 50. Would you like to go to the Rio carnival? No, it will be noisy and dangerous. But if you think about it, the more experience you’ve had in life, the more sense it makes to draw on your experience and do things you’ve done before because you’ve got more past experience to be habitual with. But habit is perfectly rational, it’s what’s called ecologically rational. Given the environment in which we’ve grown up and the information we generally have available when we make a decision which is imperfect, making this decision I’ve always done it before and therefore I will continue doing it again is not irrational, once you understand that people aren’t trying to find perfect but avoid a crap holiday. And going to places you’ve been before is a good way to avoid a terrible holiday.

Social copying

You all know that instinctive feeling when you go in a restaurant and you discover you’re the only person eating there and you’re nervous and think this is gonna be awful! That’s again really is deep instinct which if you think about it, the fact that other people do something it’s not evidence that it’s brilliant but it’s quite reliable evidence that it’s not terrible. If you buy the market leading car it may not be the best car in the marketplace, but it’s highly unlikely to be an awful car! Ok? And if you think about it, social copying, if you imagine you’re on a mountain, there’s heavy fog, you can barely see and don’t know where you are or trying to get to and stumbling around you happen on a path. You don’t know where the path goes, you don’t even know in which direction to take the path, you will still follow the path. Why? There’s only why thing you know really about that path which is that lots of people have walked down before, some of whom have had better visibility than me. Wherever that path goes, it almost certainly doesn’t go straight over a cliff. And you will instinctively follow the path by copying other people not because it’s a perfect thing to do but because it’s a non-catastrophic thing to do.

This instinct sometimes overpower us, people taking penalties in world cup finals, disproportionately tend to kick the ball left or right rather than down the middle. Not everybody, obviously you can kick it down the middle, but a few people kick it down the middle would be more likely to score than kicking it left or right because the goalkeeper dives left or right more often than he stays still. Why do people do this? It’s not because you’re more likely to score when you kick it left or right, it’s that you look less stupid when you fail.

Now a huge amount of human behaviour is driven by that kind of thing, I haven’t got time to talk about it, but in collective decision making in business decision making, good causes us defensive decision making. And it often leads people to go with very, very conservative conventional solutions not because they’re better but because there’s less blame in the event that they go wrong. Any game theory nuts here, Minmax or Maximin? Is where you choose the option where the worst case scenario is least catastrophic. No one ever got fired for buying an IBM, it’s effectively what I’m looking at, the least catastrophic. That’s why there are 4 big accounting firms, you appoint PWC if something goes wrong, people blame PWC. If you blame a small boutique accounting firm, it might actually be better for what you need done but if something goes wrong, you made an eccentric decision so now they will blame you. The huge amount of inherent conservatism in business decision making which is driven by this thing, what’s the worst that can happen? Minimize that worst case scenario.

It’s why by the way – I don’t know if anybody knows, it’s why when you book – if you ever book a flight through a PA in the office or through a travel agent, they always book you out of Heathrow rather than London City? The reason is if your PA books you from Heathrow if anything goes wrong you blame British Airways because Heathrow is a non-decision. It’s a default – you can’t ring your PA from Heathrow going ‘what the hell were you thinking booking a flight from the world’s second biggest international airport?’ Are you completely mad? London City for many people is probably a much better airport, it’s much quicker check-in, shorter distance if you were in the city or East London. No, the problem is if you book a flight from it you have actually made a decision. Heathrow isn’t a decision, a default. Book someone out of London City and if something goes wrong, they won’t blame British Airways, they might blame you.

So understanding that kind of paranoia in collective decision making is important but I would argue by the way as a marketing man, once you understand that avoid the terrible rather than seek the perfect, that’s why brand preference exists.

A brand is not a very good guarantee that something is perfect, ok? I’m in advertising and I will admit that, a brand is a very reliable guarantee of something not being terrible. Why? Because if you have a reputation that you’ve built up over 30 years, it is much more costly to you to sell a bad product than it is to someone who has no reputation to lose. You’ve got your head in the game, they haven’t. They can just reinvent themselves under a new name and make some terrible televisions somewhere else. Brand preference is a perfectly rational form of catastrophe avoidance, it’s why – you all know this, don’t you? Rationally, you will probably go online and read loads of the reviews, you look at the specs. Nonetheless you feel your money brain feels more comfortable buying something with a recognisable brand attached to it than something that has a name you’ve never heard of attached to it. Not irrational, it’s actually quite clever while you understand the information theory going on and what the person is trying to do, they’re not trying to buy the best TV in the world because the difference because good and very good TV’s is pretty futile to be honest. The difference between a good and a bad TV is massive in terms of general human misery. I had a friend who didn’t learn this lesson and they bought a TV, fantastic picture but for no apparent reason an enormous green 1 with an arrow would show up on the side of the screen. So they would enjoy the champions league and you were sitting there going why can’t we get rid of that effing arrow?

Advertising, it’s a costly means of signalling that you have faith in the product you’re selling because it would only pay if the product were widely and repeatedly popular and if you had long term intentions for it. It’s like an engagement ring, sorry to be unromantic. Upfront expense, costly proof of long term intention, ok? Because if you were just planning one night stands, you wouldn’t spend 3000 on a ring, ok? There are cheaper alternatives apparently. So what you’re seeing is our brain is hugely well attuned to understanding these things on a non-conscious level and we’re better at making judgments on these kinds of thing using the unconscious than we are using reason. I will give you an interesting example, when a few friends of our left university about 4 years after they left, we all had enough money to buy our first crappy second hand cars. We were all living in London and one of us bought a second hand car in London. We were like the salmon swimming home, we all went to the small towns where we grew up and bought a second hand car from someone vaguely known to our dad. There’s a reason that’s clever, that person might sell dodgy cars, but they won’t sell a dodgy car to the son of someone who drinks in the same pub as all their potential customers. There’s a reputational feedback loop in a small town which doesn’t exist in a large city where you go by mobile phone through a guy you meet in a car park called Dave with a paired mobile phone number. Not the same thing.

This is an interesting case, the world’s best brand in many way, certainly in the tech space, Apple TV.

How many of you have it? That’s quite a large proportion and you’re not representative but it’s not expensive, it’s a very good product but the thing is until recently Apple never advertised it. They sold it but they never put any scarce resource like advertising media budget or Steve’s time on stage to talk about it. And the consumers natural influence was they’re probably not that serious about this product, it might be a hobby and I could buy it and end up getting betamaxed and find out they don’t really support the platform.

When you advertise things, it’s the difference between saying your horse is gonna win on Saturday and betting. One of them is cheap talk, the other one is a costly enhanced reliable signal cause there is a cost attached to you being wrong. And we instinctively seem to discern the difference between costly signals and cheap signals. A flower when you think about it, nature has a very large advertising budget. A flower is simply a weed with a marketing budget. And the interesting thing about this, I actually surmised this and thought there must be a bit more going on than just make the flower really noticeable so bees go and pollenate it. And flowers produce a lot of smells, but there’s a particular smell that flowers produce which they can only produce if they have a lot of resources available for producing pollen. So that’s a costly signal, you can’t bullshit with that smell because you can’t produce a lot of the smell and bugger and pollen. You can’t cheat the bees by only reducing 6 grades of pollen and a ton of smell, because of their ability to produce the smell is correlated with their ability to produce the pollen and bees have learned that’s the reliable signal and that’s the smell they prefer because that’s the smell you can’t fake. And I think what’s fascinating is in nature there are all these things going on in terms of signalling which biologists understand advertising much better than economists do, they understand these things about the costly signal.

In other things you can do, if you look at how my mom bought a car- this is understanding calibration again. Perfectly good way of not buying a terrible car without knowing anything about cars? Buy a car from someone you trust. You don’t have to believe in god, but if you buy a car from a vicar, you know that he’s unlikely and will suffer a lot of reputational damage if you find out that the car is in fact 2 cars welded together and the gear box was full of sawdust. That costs the vicar quite a lot. So buying from someone with a reputation is a good way of not achieving perfection but avoiding disaster.

Imagine this, you turn up at a house to buy a second hand car. You look at it outside and think it’s pretty nice and prepared to pay 5000 for it so you ring on the doorbell on the person who owns the car. In what universe is the door opened by a vicar? Now the car doesn’t change, in one universe the door is opened by a vicar. You will raise the amount you’re prepared to pay by 500-1000. In the second universe the door is opened by a man in his underpants. Now the car hasn’t changed but immediately the amount you’re prepared to pay has dropped. True? Why? Because someone who opens the door in their underpants is not vulnerable to shame. Ok? And your brains is doing that calculation, you would automatically go shit, I was quite keen on that car and now I’ve seen his underpants, I’m not interested anymore. The car hasn’t changed where automatically we’re assembling these little bits of information which are reliable information that help us avoid disaster.

By the way, if you satisfice rather than maximizing, you’re happier. People who go I’m gonna buy a pretty good holiday which definitely isn’t terrible! They end up much happier than the people who spend hours trying to have the perfect holiday. I suspect it’s because the perfect holiday is great when it’s perfect but when it goes wrong it’s terrible. I imagine it’s been – if you’re choosing a wife, there’s what you might call a higher optimality and high risk option which means New Yorker wouldn’t it? [laughter] And then there’s pretty good and not rubbish, west country, you know what I mean? A Bristolian, maybe Welsh actually. Any female from a culture with really low expectations of male behaviour basically [laughter] but the point about the brain – this is Jonathan Haidt model. It’s a rider on an elephant and the rider is the conscious part of the brain, that uses reason. It’s slow, deliberative, it can only process so much information at one time and a highly constrained bandwidth, the elephant is basically all the unconscious mental mechanisms which go on on autopilot.

What’s so significant about that analogy is that the rider kind of thinks he’s in control of the elephant. He might be able to nudge the elephant’s attention here or there or direct it to look at something or encourage or discourage it a bit, fundamentally thought the rider can’t get the elephant to do anything it doesn’t want to do. The other important point about the analogy is the elephant doesn’t talk, doesn’t generate reasons like our conscious brain, it influences our behaviour by generating emotions. Emotions and heuristics which are the stock and trade of the elephant brain when you think about them, have a huge advantage over reasons in an evolutionary sense which is they can be inherited whereas reasons have to be taught.

My own – that’s strange – analogy for understanding the human brain is the power of the rational brain is effectively over decision making will stick with this dual decision making model of the brain, it’s like a colour blind man buying a sofa with his wife. Which is if you’re a bloke who goes shopping with your wife, you might be given a casting vote very occasionally, you might be able to direct attention by saying have you seen these sofas over there? There’s a bit you can do but you can never end up with a sofa that your wife doesn’t like. There’s not a chance in hell of that ever happening, when you end up putting it in the car or getting it delivered, whatever your opinion and involvement in the decision is, if your wife isn’t happy with the sofa it isn’t happening. True? There is no way in hell you could ever impose a man sofa on your female partner. A man’s sofa would probably have a beer fridge built in and a laptop stand. What else would you do? You’d have a gadgets – the sky remote built into the armrest. You will never get that home. So that’s the relative power at play in the brain in terms of decision making. Ok, conscious rationality can play its part, but fundamentally it’s only when you’re emotionally happy with something that you can do it.

The problem with this is the elephant trumps its feet and generates emotional reactions, it doesn’t generate reasons and yet most business decisions are taking by using what I call the dodgy pair of binoculars, they are neo-classical there and there’s market research here. The problem with market research is by definition you’re talking to the rider, not the elephant. It’s like trying to design sofas by talking to husbands, you’re not talking to the decision maker, if you get my drift. Now the point is that neo-classic economics has a fundamental problem and there’s a lot of things which are hugely important to the elephant which standard theory of human behaviour doesn’t understand.

This is David Rock, a neuroscientist, and these are his 5 status, certainty, autonomy, relatedness and fairness. I haven’t got the time to talk about all of them. Autonomy matters for its own sake, ok? that’s probably quite big behind the Brexit movement. It’s why the movement has difficulty actually voicing in rational terms their emotional feeling because they basically just go I prefer it when we can do our own thing, but that’s not a quantifiable argument it’s an emotional argument.

Fairness, though it doesn’t just apply to human instinct, the higher primates, particularly those that operate in social groups also have an innate sense of what’s fair.

This fundamental problem is when we talk to the writer, we get a post-rationalised explanation of our past behaviour and a kind of pre-rationalised predicting of our future behaviour which doesn’t account for any of the mental processes which go on in those instinctive elephant parts of the brain which are a pain to introspection. They processes is like you saw in the McGurk effect where without you being aware of it, your brain was presenting information to you in a more digestible form. And so what I always find interesting is when you look at really successful products, in business it’s considered much more acceptable to win an argument or battle, consumer or marketing battle, through reason rather than emotion.

If you take the Dyson. James who is an engineer believes I think, although he’s starting to doubt this, but he believed the reasons the Dyson’s were superior and very popular was because they didn’t lose suction and they were bagless. No loss of suction as the container filled up and that they also didn’t need bags. I would argue baglessness is a disadvantage of the vacuum cleaner cause it makes them difficult to empty but we will park that obvious objection. I think that’s right, I don’t think there’s things would have sold well if they were rubbish, I’m not suggesting that for a second however a very large part of the elephant appeal of the Dyson apart from the part that it looks incredibly cool, the elephant brain likes the look of that thing and it’s inherently more satisfying when you can watch the dirt accumulate in the container. The monkey brain just really likes doing something when there’s a visible source of feedback. Now what I’m not saying is he could have sold crap vacuum cleaners that were transparent, not at that price. Cause they’re pretty pricey. What I would say is had he made the outer casing opaque and beige I don’t think he would have sold many. I think that’s as much an elephant purchase as a rational one.

Uber is very largely elephant driven. The elephant brain hates uncertainty and will do everything it can to resolve dilemmas the best piece of marketing advice I gave my entire life was to say to British Airways the worst thing you can put beside a departure board is the word delayed. That’s just crashed is worse but we’ll pardon that one [laughter]. I said make up a time that’s fine. If you don’t know if it will be delayed 30-60 minutes just say delayed 50 minutes. We’re cool with that. We got bad information but it’s information we can act on. I can go and have a meal, I can go to the lounge and get my laptop out, ring people and say we’ll be an hour late. It’s ok, I know what to do. But the single word delayed you might as well just put be miserable.

The single best thing London Transport did to improve passenger satisfaction on the tube per pound spent wasn’t faster, more frequent late running trains, it was the display boards on the platforms. Doesn’t change reality at all but we’d rather wait 8 minutes for a tube knowing it’s 8 minutes than wait 5 minutes for a tube in a state of not knowing and uncertainty. We inherently really hate uncertainty. Uber has 2 utterly brilliant mind hacks, it’s their 10-15-20% better than other cab firms on a variety of measures like price and speed of turning up and cars available and so forth, but being 15-20% better doesn’t generally change human behaviour much. It’s a bit much like Heinz tomato ketchup which you can’t beat because any improvement you make is counteracted by the fact that it tastes unfamiliar and it’s very difficult to be better than Heinz because any gain you enjoy in taste is offset by the dislike with the slight weirdness and I think in many cases.

I don’t think Uber would have succeeded if you’d still have to make a phone call to a call centre to book an Uber. The real mind hacks here are before you book you can see how many taxis are around so you know whether you need to book it in advance or you can do it at the last minute, elephant likes that. The second thing is previously when you booked a taxi you entered the twilight zone of uncertainty. Is he here yet? Why isn’t he here? They said 10 minutes, can you go out and stand in the rain? It was a total pain in the ass between booking a cab and climbing into the damn thing. Cognitively it was just awful! There’s no objective measure for the pain of uncertainty but for time and to the human brain, that is the real clincher. Here you just look at the map and say he’s stuck at those traffic lights, I will have another pint. Or if you’re weird like me – does anybody else do this with Uber, where you actually try and time your emergence onto the pavement to coincide exactly with the arrival of the car so you can re-enact the final scene of the usual suspects where Kaiser Soze climbs into that Jag, right? Cause that feels really cool!

The other thing – the elephant just loves this shit – is the means of payment. If you had to pay cash and fart around with receipts it would lose a large amount of its appeal. The physical and mental pain of payment is massively reduced if you can separate payment from consumption. If you can make it happen at one remove and it’s invisible, it feels less expensive than paying with chip and pin, they feel less expensive than paying with cash. Don’t bother making a product that’s cheaper, just find easy ways for them to pay for them. I mean the Starbucks – there was a prepaid card, where you spend the money on that card, it doesn’t feel like real money does it? You spend 4 pounds on a prepaid card and it feels like spending 2.80 pounds. Cause the amount of paying the money was pre-committed, nothing is actually happening here, I’m not getting any poorer and I’m just effectively through a process of transmutation turning a bit of imaginary money into coffee. It’s not the same thing. But did you know apparently 500 million sitting on those prepaid cards and apps at any one time for Starbucks.

Eurostar you probably all heard me say this, don’t spend 6 billion reducing the journey time by 40 minutes. Really extravagant, I said put Wi-Fi on the trains. But one of them is a macho engineering solution, makes the train faster with objective metrics like speed, duration and the other one is a soft measure which is how enjoyable is your journey? The point I would argue about Eurostar is there’s no point trying to be fast because you’ll never be as fast as an aircraft. The job of Eurostar is not to be fast, it’s to be better than the plane. I would argue the way to compete with aircraft is not in speed because they have a considerable advantage in speed because they are technically known as very fast. Now, the great advantage of a train over a plane is that you sit on your ass for 3 hours. A 1 hour flight has 1 hour of dicking around at either end. Therefore the killing advantage of the train is open laptop, log on, use it for 3 hours, get off in the middle of the city. The plane cannot compete with that under any circumstances. Speed is something where to some extent you played your weakest suite against planes. I did say it and you spend 6 billion knocking 40 minutes on the journey time. You could have spent 1 billion and hired all the world’s supermodels, got them to walk up and down the train giving passengers drinks. You would save 5 billion and people would have asked for the trains to be slowed down.

But there’s a very strange thing which you all come across all the time. Solving problems psychologically is viewed as cheating. Look at how medicine treats the placebo effect. The guy I will show you at the end who explains why the placebo effect works, but doctors see it as macho to come up with elaborate drugs whereas creating an environment which encourages the body to heal itself is regarded as cheating. It’s just weird! I don’t really know why this is. I went to the government and I said you don’t need to spend 60 billion on Hispeed 2. I could get greater capacity on the trains to Birmingham and Manchester and reduce the journey time by 40 minutes by spending about 100k rather than the 60 billion. They said that is stupid, I said no. Every time I go to Manchester I buy an advanced ticket. We all do, otherwise you’re mad! Nobody buys a full fare ticket. The problem is that is requires you to travel on a specific train and therefore you’re terrified of missing that train which means your ticket becomes void and you have to pay a full ticket so you take an insane amount of margin of error and you spend 45 minutes hanging around at Burger King eating their not so healthy breakfast. All you need is an app that says I’m here at Houston right now and Virgin Trains replies there’s a train leaving in 5 minutes time where seat H7 is empty. Pay us 5 quid and you can sit there! Now, I won’t get into the details of yield management, reduces journey time by 40 minutes. If you allow people to travel on early unoccupied seats, that’s good yield management practice, because seats are perishable good and they will let you travel on an early flight which isn’t fully booked on the grounds that it gives them a chance to sell your seat sometime later on. You can’t allow people to do it on later trains and you can’t guarantee that people will go on an earlier train but if you made that system operational it would improve yield management, increase capacity and reduce journey time at what is literally the cost of a couple pairs of signals on HiSpeed 2 and because it’s psychological rather than infrastructural it’s seen as cheating.

The other thing is that psychology can be used to solve problems in a way that pure logic can’t. once you understand that the human brain hates uncertainty, a lot of people would say this problem which is you either reinforce the cockpit or you don’t, in one case there’s the risk of pilot suicide and the other one there’s terrorist takeover. Once you understand a bit about the human brain, what you realise is that you can solve the problem in both directions, you just create uncertainty. So some of the planes are fully reinforced, there’s a code given to the stewardess which sometimes works and sometimes doesn’t. If you put a bit of randomness in, nobody is gonna hijack a plane and crash it into a hill if they think there’s a 40% chance that their plan won’t succeed. You just have to build enough doubt to stop people doing things. This is why the whole business of compliance is more and more focused on psychology than on what you might call rulemaking and box ticking and procedures and rules.

Now I mentioned this about the path dependency question. There is a weird thing about the human brain which is that when questions, we’ve evolved as a species to make the best use of information that we can. In real-time, because that’s what evolution needs you to do, make pretty good non-disastrous real-time decisions using whatever information is to hand. The result of this is when the information is very bad presented to us, we don’t notice. Now I will give you an extraordinary example of this, has anyone been on the M20 heading towards London, needed to turn off for the M25 and missed the turn off? Anybody made that – the two people in the back there. I had one person in Johannesburg who’s missed the turning off the M20 for the M25 and I’ve done this 6 times out of 12 and after the 6th time I go it’s not that you’re a total idiot, there’s something wrong here. So I go and google streetview and this is the first sing says. It’s not clear because nobody knows if swanley is north or south. It’s a very, very bad sign but you go it doesn’t matter cause I can see this other sign here so I will wait until I get to the second sign, I will read it and I will make a decision then this is what happens when you approach the next sign. Before you can read what it says – so for all the value it has that sign might say you are an idiot! Now the strange thing is it took me 5 or 10 minutes on google street view to realise how stupid that sign way. The highway agency was completely unaware with how stupid it is. It’s actually not stupid but dangerous because at this point you are tempted to swerve over the solid line which is completely illegal in order having to avoid going back to the M25.

The weird thing is when information is presented to us in a crap order, we don’t notice and the people who design questionnaires don’t notice. So this is a classic case of where we noticed there was something really dumb here. The standard airline booking form on the web says where you want to go, when and what class of travel do you want? Now for business travel that’s ok, generally business travel there’s a specific place and date I have to be there, it’s very rare someone at the office says sometime next month we’d like you to go somewhere sunny, generally I don’t get that kind of brief. We’ll park that for now.

As a business traveler, what flight class you want is not a stupid question because your company tells you what flight class you’re allowed – does anyone realise how stupid that is as a consumer question? Until I know the difference in price it’s impossible for me to ask the question. It’s exactly like the road sign, you’ve asked me a question to a point, you forced me to make a decision at a point where I don’t have enough information to make that decision. So if you think about it, you can’t pay – paying 1000 for business class if premium economy were 800, paying 800 for premium economy were fine if economy would be 350 and wouldn’t be ok if it’s 300. It’s like going into a shop and asking which one do you want? The expensive or cheap one? What’s the difference in price? Not telling! It’s a dumb questions and airlines have asked millions of people that. Now it’s halfway fixed with an airline company of ours, even if you ask for economy, we show them the premium economy and the business class price, even if they requested economy. That also helps because it makes economy look cheap as well. So it works on an expresso principle as well as that – that brought in an 80 million of incremental premium revenue the following year. Previously people said I have to ask for economy and they found an ok price, they didn’t bother to check the other price and booked economy. In many cases, premium economy which is a much more profitable cabin by the way, nobody even bothered to look. The second people saw the comparatively it wasn’t that much more expensive, I think it’s important and a ratio that we pay, we don’t have a set amount we’re prepared to pay for one class versus another.

We don’t also realise when the way in which choices are presented to us are totally deforming so you can manipulate people without realising. When you go to the restaurant, the restaurant wants to sell you wine. Wine is much more profitable than any other alcoholic drink cause it doesn’t have a known price anchor. You can’t charge people 30 quid for a glass of Johnny Walker Red cause they know all that cost in the shop, but you can buy a bottle of chateau of obscure for 5 euros, charge 38 at the table and everyone goes marvellous! Hint of black current! Total bullshit! Now what happens when you go to that restaurant? They put wine glasses on the table, we – if you take them off the table, far more people drink beer and fewer people drink wine. Then the next thing that happens is they bring you a drinks lists and it’s just not called the drinks list. It’s called the wine list, so you’re being nudged again. And the wine list has a choice of architecture which is basically 8 pages of an insane range of different wines and one page at the back with all the deviants and perverts who actually want to drink something that comes from a culture that’s mastered the technology required for distillation and brewing rather than just getting some peasants to stomp around on some grapes for a bit [laughter]. Big bias there, ok I grant that one! And then finally, there’s a bit of genius which I only noticed a few months ago, they only hand out one wine list. So there’s one guy sitting there with the wine list, now there’s only one alcoholic drink that one person can order for anybody else. You could go tequila slammer all around but it could be frowned upon in bourgeois circles. So what happens with the guy with the wine list? He has it and looks to the table and says red or white? And at that point it’s game over for the beer drinkers and gin drinkers and everybody else and you basically been completely manipulated just as badly as you were on the M20 motorway but again you don’t realise. We have to be really careful about this!

Now I’m probably running a bit out of time now. Shit! Ok! [laughter]. Right, I will end on a quick note. This is important because we don’t realise how our elephant is making decisions, because we don’t understand elephant logic and we’ve never really codified it, we don’t understand how elephants make decisions. But it’s hugely important because there are loads of things which affect. Those of you who are my age can remember when most families there were 3 kinds of families in the UK when I was a kid. Really rich people who had more than one car, people that had a car and those that didn’t have a car. And nearly everybody who had a car had a saloon car, the Ford Cortina would be the absolute standard. The reason nobody no one has a Ford Cortina – how many people have a conventional saloon car? Nobody has a saloon car and part of the reason for this is when you have 2 cars, nobody buys 2 saloon cars, you go more extreme. You have an off roader and a small Mini or an SUV and something else. When you get one person choosing 2 things, it’s completely different from 2 people choosing one thing each.

Tesco had a huge dominance when everyone did a £120 weekly shop cause you had to shop in the middle of the market when you were doing one massive shop cause if you bought everything from M&S you’d go bankrupt and from Lidl you would feel you were under house arrest in East Germany, it just felt miserable! So you want to go to Tesco and that was a happy compromise. Tesco taught people that small shops can actually be good. So once they discovered that Tesco mantra they started doing 4 £30 shops instead of 1 £120 shop. When you do 4 of them, you don’t do all of them at Tesco you do one of Waitrose, the farmer’s market and one of them at Tesco. So when people shop big then frequent Tesco has dominance, when people do small and more frequent actually the market broadens massively.

And we do that instinctively, ok? nobody – if you gave one person 400,000 to buy 2 houses, it would be totally different to giving 2 people 400,000 to buy one house. What they come up with – and yet we’re totally unaware of this decision making process which is a pity because if you want to recruit people and genuine diversity in your workforce, hire people in groups! The only reason I got my job there and they had 4 jobs and they decided on the other 3 people and let’s take a punt on the weirdo! Now you don’t take a punt on the weirdo if you’re only hiring one person. Got that? If you’re hiring 5 or 6 people, your variance will go really quite high. If you’re hiring only one person, defensive decision making takes over and you hire someone really close to the expected norm, the PWC of the internal appointments. If you hire 4-6 people, the degree to which you broaden your repertoire is massive and that happens without quotas or legislation, it’s just an automatic process in the human brain. And what’s so strange about these processes is because our conscious brain is in denial about them and wants to pretend that its in control, we don’t even know that they exist.


This is why it’s a really – it’s my last slide, I promise! This is a man called Nicholas Humphrey and we talked about we think that medicine is good and placebos are bad and we think building new railway tracks is good while putting Wi-Fi on trains is cheating. We seem to have some weird aversions to solutions that acknowledge they exist in the unconscious, ok? Dyson would love to stand there and talk about superior suction if you actually said my monkey brain loves the fact that you can see the dirt accumulating inside, we really don’t like that as a solution or he doesn’t.

Nicholas Humphrey has a theory about the placebo effect which is that inside the unconscious or in parts of the brain in which we have no control there’s something you might call a central governor. It’s a bit like the internal accounting system of the human body which decides at any one time how much in terms of resources can be dedicated to different things and digestion, fighting infection, recovery and repair. All those different things which have a call on our sources of energy – they’re effectively allocated resources, according to the context by a central governor which is the product of evolution. Imagine you sprained your ankle and it hurts not because it actually hurts, your pain is there to stop you from encouraging you to use the ankle so that it heals more quickly. But if while you had the hurty ankle and you’re attacked by a tiger you would immediately forget the ankle hurt and just run like hell because the central government decided quick change in priorities, ankle repair, low priority – high priority, ok? So there’s function with very quickly switch off ankle pain and devote loads of resources to flight.

Now his belief is the placebo effect works because when you take unpleasant tasting pills, when your family and friends come around to make a fuss of you, when the days are long so it feels like summer, hamsters – by the way, this is why you get more colds in the winter, because your internal governor decides in the winter it needs to conserve resources more towards keeping warm so it can’t devote the same resources to fighting a cold. Hamsters actually, if you hack hamsters and give them a big day and they think it’s summer, they actually don’t get ill, ok? Now the interesting thing about this is if this is true, my grandfather was a doctor and he said that before penicillin came along, most of what you did as a doctor was psychological. If this is true, it’s hugely important! And maybe billions should be spent each year to work out how we can create the stimuli that work on the unconscious to basically say – he says that you know, our governor is calibrated to the savannah, not to the modern world where we don’t face the risk of starvation. The placebo is a way of hacking away the unconscious to basically say you could afford to give a lot of resources to getting better.

Now I’d always thought this was an annoying thing as a marketer that people would rather say improve wagamama’s food delivery rather than coming up with a really clever sentence that solves the problem on a psychological level. I always thought that was a marketing peculiarity and then I met this guy and realised that thousands of people die because we’d rather spend money on developing a drug that probably, then just developing psychological hacks that make the unconscious feel that now is the time to get better. This is actually quite an important distinction so that’s me, that’s the reading list and thanks very much indeed [clapping].

Find out more about BoS

Get details about our next conference, subscribe to our newsletter, and watch more of the great BoS Talks you hear so much about.

Rory Sutherland

Rory Sutherland, Ogilvy
Rory Sutherland

Rory is Vice Chairman at O&M Group UK, an attractively vague job title which has allowed him to co-found Ogilvychange, a behavioral science practice. You should watch this talk if you want to change your perspective on product and problems.

Rory co-heads a team of psychology graduates who look for “butterfly effects” in consumer behaviour – these are the very small contextual changes which can have enormous effects on the decisions people make – for instance tripling the sales rate of a call centre by adding just a few sentences to the script. Put another way, lots of agencies will talk about “bought, owned and earned” media: we also look for “invented media” and “discovered media”: seeking out those unexpected (and inexpensive) nudges that transform the way that people think and act.

More From Rory.

Learn how great SaaS & software companies are run

We produce exceptional conferences & content that will help you build better products & companies.

Join our friendly list for event updates, ideas & inspiration.

Unsubscribe any time. We will never sell your email address. It is yours.