Why You Keep Falling for Ads — Explained by Rory Sutherland

If you are famous, you are more likely to get lucky than if you're obscure. If you are famous, your customers find you. If you're obscure, you have to spend your whole life finding your customer.

Today, we have the famous Rory Sutherland, vice chairman of Ogilvy and one of the most influential thinkers in behavioral science and marketing. The fundamental thing that distinguishes a good CEO from a bad CEO is that a good CEO acknowledges that business is probabilistic. It's not deterministic. He's a champion of unconventional ideas. He explores how human psychology shapes decision-making and challenges traditional economic thinking. I don't think you should be a marketer without actually having enough respect for the financials of a business. Sometimes you've got to piss off your existing customer and find new customers somewhere else because a renowned speaker, columnist, author of Alchemy, Rory brings a unique perspective on how creativity and behavioral insights drive commercial success. You need fundamentally different accounting principles for innovation and marketing. Exploitation funds the exploration, but the exploration informs the exploitation.

[Music] So today's episode is, of course, brought to you by two fantastic organizations. The first, Studio Space, a leading on-demand marketing agency platform. This enables marketers to browse and brief agencies, receive proposals, and start work all in one place. It's literally like having your own agency network at your fingertips without the hassle, as they take care of everything from contracting, invoicing, procurement to legal. So check out Studio Space today. Our other sponsor, of course, is the Marketing Skills Trust, a charity whose mission is to make the marketing industry a better place, providing grants to help create a more equitable industry and help us to level up. The Marketing Skills Trust has been instrumental in championing a more inclusive, diverse, and open marketing industry. You can find out more by going to their website.

Okay, let's get on with the show. So Rory, fabulous to have you on the show, and um, I don't know if you were aware of this, but you're the only person we've ever had back for a rerun. The only—I, I had to be really careful not to repeat myself, so uh, then I'll do my best. We'll navigate through that. No, it's honestly a privilege to have you on the show. Let's, let's get by started by asking how's, how's 2025? What are you learning about the world in 2025?

Um, crikey, in some ways I'm getting—I, I have smidges of optimism about effectively the rising importance of psychology and indeed marketing within business decision-making. I see pockets of optimism; the overall prognosis I still—I'm still fairly depressed by, which is—I mean, I'm intrigued also by the relative imbalance of what we talk about. But this is not to say that AI is not important, by the way. I think it is very important, and I think uh, in 5 to 10 years' time, its significance will be immense. It is, on the other hand, hard to predict and currently hard to experiment with because we don't yet know the direction of travel fully. On the other hand, I think there's too little discussion about things like, for example, flexible working or video conferencing, which strike me as very, very important technologies that never have a hype cycle. Uh, various technologies which, because they play out gradually and irreversibly over a decade or so. The case of the dishwasher—we still haven't reached 50% dishwasher penetration in the UK. You know, these things are—it's astonishing. These things are actually weirdly slow, but they're kind of irrevocable. Electrification being another thing. We never talked about LED lighting. Now, in, if I get this right, I think in 2010, 1% of lights sold were LED lights. By 2030, it'll basically be 100%. That the invention of the blue LED, which made white LED lighting possible, I think is equivalent to like removing a quarter of the world's road traffic on a like-for-like comparison in terms of carbon emissions because of the huge efficiency gain. What's weird is nobody talks about it. And so it does strike me that we're living in a world where attention has become a kind of winner-takes-all. Uh, it has a winner-takes-all aspect to it, which, as a consequence, means our attention is imbalanced. By the way, the things we talk about a lot—I mean, AI—one optimistic interpretation of AI is: do large language models rebalance discussion from the numerical to the verbal? That's one possible benign consequence; that because you can—because effectively numbers were much more amendable to computation than words were—the whole business world suffering from massive quantification bias where we deemed important those things that would fit on a spreadsheet and pretended that everything else was too fluffy to care about. So, you know, is there—you know, does this actually lead to, you know, an interesting kind of rebalancing? It strikes me as really weird, by the way, that we're completely happy taking advice from a black box when it's a machine. But the human brain has its own black boxes, which are things like intuition and instinct, which in a business context you're not allowed to use at all. Okay. You will never win an argument by saying my instinct is that this will work better. You have to come up with data. Okay. Strangely, if the instinctive response is generated by silicon rather than carbon, we're all treating it with extraordinary sort of deference and uh, and uh, credibility.

You know, Rory, I find this—now we're going down the rabbit hole of AI in, in that, in that sense. I, I find this an interesting thought because when you look at—you know, you kind of alluded to the fact that humans are very unable to kind of think about compounding and exponential growth and and how it sort of comes together—and AI is pipped to kind of double—double faster than Moore's Law in that sense—in that it's going to, you know, it's going to increase in its computational capabilities, doubling at its speed year on year, to the point where they become super intelligent, as we know possibly, and then thereafter we look like a spare part as opposed to the guys in charge—and, and I wonder then therefore what's your perspective in 5 years' time if we get to that sort of, that moment and um, and how do we then see the world of marketing, advertising, creativity—do we get run by the machines at that moment? What's your perspective?

I think it'll play out in different ways. I mean, those of you who see me speak know that I'm a weird defender of chains. It's not fashionable to say, "Isn't Travelodge brilliant?" Or, "I think Starbucks or Costa are wonderful," because they're large chains, and I always defend chains on a very simple thing, uh, which is uh, they don't necessarily raise the ceiling, but they massively raise the floor. Okay. So I grew up—you did as well, probably—in an age where 30% of hotels were just simply appalling. Right? I mean, when you booked into a hotel, there was an element of just a gamble. Um, and the great thing about chains, whether it's coffee or hotel rooms, is that you can't survive being bad anymore. And that has benefits for a whole category. You know, it, it raises expectations in a whole category. A part of AI will work in that way of raising—raising the floor, not raising the ceiling. I think that, you know, you might argue in fields like medicine—even if all it does is provide a checklist or an intelligent list of questions—um, it will prevent people doing things that are utterly idiotic, and it will raise—arguably you might say—it will raise doctors at least to the level of an average or pretty good doctor. That's one—that's one reason to be optimistic. There are reasons to be pessimistic, uh, which are, by the way, the reasons to be pessimistic are actually psychological. They're not technological—are in many ways. Second thing would be it might—you might allow humans still to perform the last mile. But if it starts off with a better question, or it gets you to a better starting point or a more interesting starting point, uh, that stimulus might then hand over to a human who gets to a much more interesting endpoint that you might either never have reached with AI or which would have taken you much faster. Um, and in some ways I think—the way I understand this, the way I look at this—because I think the problems are psychological, not technological to a large extent—I look at things that are analogous to AI, which actually are technically AI-like—a satnav. Okay. Now, sometimes your satnav can show elements of genius; in other words, it will give you a route that you genuinely wouldn't—I pride myself on creative thinking ability—I was a very early satnav owner, and I had a meeting. I live in Kent. I had a meeting with EasyJet, and there was some catastrophic traffic disaster, and I like to think of myself as an imaginative person, but the solution it came up with—which was go the wrong way around the M25—I wouldn't have actually thought of myself. I was agonizing about how to avoid, you know, bits of the southwest quadrant of the M25, and it just said, yeah, basically head for Dartford. And so it kept me—brilliant. Um, sometimes though you ignore your satnav because it doesn't—this is called the alignment problem—it doesn't fully know what you're trying to do. You may want the scenic route. Okay. Uh, you may be trying to catch a plane where the most important thing is not average journey time but low variance. You know, you take the motorway, it's faster on average, but if something goes wrong, you're stuck for an hour and a half. So AI as what you might call an advisor or stimulant—bluntly put, I'm very, very optimistic about. What worries me is in an institutional setting when you're a doctor and the AI recommends something. One thing that humans are very good at is going, "I see what the AI is trying to do, but this time it's different," or, "I see why the AI thinks this, but I think this is a special case." And so sometimes you'll look at your satnav and go, "Yeah, nice, nice try, mate, but actually it's dark. It's raining. There's no way I'm going to save 3 minutes by driving down back roads. Sod you. I'm going straight to the motorway." Okay? Because this—this is—this comes down to the in-between thing, by the way, of effectively what the left hemisphere of the brain does in humans and the right hemisphere of the brain—the left hemisphere effectively optimizes for a narrow context, and the right hemisphere considers the wider context within which the decision is being made. Okay. And so what would worry me in an institutional setting, not in an individual setting—I give AI suggests holidays to you. You—you go—what the hell—everyone goes, "Oh, I never thought of going there." In that sense, it can be an extraordinary spur to creative, variant decision-making. AI property searches strike me as potentially fascinating. Okay. Um, but in an institutional setting, something terrifies me, which is a doctor who goes, "The AI is telling me to operate. I don't think we should operate. But now I have an asymmetry, which is if I obey what the AI recommends, I can't be sued and I can't be blamed. Whereas if I deviate from this recommendation, I'm now exposed to blame, litigation, and everything else." And that does strike me as a potential problem. The other thing that would strike me as a problem, and here the analogy isn't really satnav, it's probably speed cameras, is when AI stops suggesting things and starts enacting them. Um, that, that strikes me as a fundamental difference. I don't think personally—okay, I think speed cameras are actually unethical because I think if you're going to fine someone and give someone three penalty points, at the very least a panel of three people should look at 90 seconds of video and consider wider context. Like, it's the A—okay, it's the A13 at 2 in the morning. The reason there's a 40 limit on the A13 uh is basically to prevent congestion. It's not a road safety issue. Okay. Now, at 3:00 in the morning, driving on a dual carriageway at 40 mph feels weird. You think you're going to get hit from behind. Uh, in some cases, the—if you look at a camera that's catching disproportionately many people, you may say, "Hold on, this could be a problem with the people. It could be a problem with the location or the calibration of the camera. It could be a problem with the road signage, because let's face it, you could design a speed camera to catch everybody if you wanted to." Yes. So this—in other words, this wider contextualization of what's going on, which is a human technique, which is, of course, an AI is blind to absolutely everything outside the model within which it's been taught. Okay. By the way, so that's another point. The third thing which interests me is what you might call agentic AI uh or AI which effectively changes the interface with which consumers interact with businesses and brands. And the reason that's important is I think the lesson of the last 30, 40 years is there are two ways to disrupt a category. Okay? One of which is to go head-to-head with the category leader or the dominant player. And you can do that, but you might well fail, and it takes bloody ages and it costs a fortune. The other way to innovate—in marketing terms—is to change the context or channel or place or time or moment at which the consumer interacts with you. Direct Line. Okay. You didn't actually tackle insurance companies head-to-head. You offered people—instead of going to an insurance broker, you picked up the phone—and then later on you went to—Okay. The Argos catalog, okay, was a disruptive retail idea because rather than people going into a shop and asking about toasters, they discovered an interesting new toaster while sitting on the toilet and went straight to Argos. Now, what—what I would say is that when the inter—when the default interface changes, all the normal rules of—you know, the normal hierarchy of the category gets reset to zero to an extent, and there are cases like Britain's banks—the high street banks—I mean, they literally spend hundreds of millions on their apps because they realize that if their app is significantly worse than the Monzo app, okay, they've got their customers leaving. Now, what they have to be is not necessarily better, but they have to be within the threshold of tolerance where people don't leave. And so, the threat of disruption through a change in interface is always there. Now, what happens if fundamentally this—this, you know, has—um—Apple Vision Pro become a massive runaway success? There'd be a load of brands asking themselves questions about that. Okay? What happens in effectively in 3D computing? How will people make choices about what to buy, where to shop, what insurance provider? Similarly, you know—I mean, if you think about it, you know, EasyJet's disruption was actually putting the phone number on the plane. That was the disruptive moment. I mean, the internet came as a massive gift because it effectively fell into their lap. But the whole thing was: don't go to a travel agent when you want a weekend break in Spitsbergen or whatever. Um, pick up the phone. If voice, for example, becomes—you know, there's a new Amazon upgraded Alexa service which is waiting in the wings. If voice becomes a major part of the interaction, um, all the normal rules—you know, you—in other words, you can build up an extraordinary brand advantage in one channel only for channel shift to effectively render what you do—what you've done today—more or less irrelevant.

So fascinating words. Just picking up on channel shift—in between times since we last spoke, you have—I think the term is—you've blown up on TikTok. I had this strange experience where one of my children said to me, "Have you—have you seen this Rory Sutherland guy?" Um, so, so tell us about how that happened.

What—totally accidental, by the way, although it is—let me explain one of the things. So someone from Sky got in touch with me about the Guinness thing—about splitting the G. Okay. And they said, "Do you think Guinness came up with that?" I said, "It's not impossible." Okay. But equally, there was always a theory that Unilever or P&G had created dressed-down Fridays, and that they deliberately and subtly promulgated this because if people didn't wear suits on Friday, but they wore chinos or polo shirts, they did more laundry and less dry cleaning. Okay? And that was always the kind of—that was the Kennedy conspiracy theory around dress-down Fridays. Now, most of these things happen by accident. Okay. So, in that sense, the marketer is not responsible. However, the more famous you are to begin with, the more likely you are to have lucky accidents happen to you. So there's a whole part of marketing spend and, and, and brand expenditure which I would argue is probabilistic. In other words, if I spend more on—there was no chance there was ever going to be a craze for people buying a pint of Murphy's and splitting the G. Okay. Right. Yes. Now, in the same way, when those lucky things happen, you know, a parallel example would be the Tim Tam Slam. If you're Australian, have you come across this? It's a biscuit, a bit like a Penguin biscuit. And the Tim Tam Slam is where you bite off opposite corners of the biscuit and then suck tea through it. Okay. So, you get a kind of chocolatey tea with a bit of caffeine. Now, it's possible Tim Tam invented that. Okay. I think he was just, you know, so—but it's more likely to happen to you if you're a big, famous brand than if you're a small, less-known brand, I would argue. And there's a very large part of advertising where we ought to be honest about this because we've gone into clients and finance. It's all about accountability. It's about measurement. We prove everything we do. Okay. But a large part of marketing is probabilistic. And I think we should have been honest enough and said, "Look, mate, it's a casino, but it's a casino with really good odds." You know, it's the unusual kind of casino where the odds are stacked in your favor. If you are famous, you are more likely to get lucky than if you're obscure. Similarly, if you are famous, um, your customers find you. If you're obscure, you have to spend your whole life finding your customers. Okay? There's a kind of inflection point at which—you know, I, I, I noticed this, which is—I, you know, I don't have to look for speaking engagements anymore because of this TikTok thing—in fact, if anything, you know, the biggest task is learning to say no to enough—and so what happened here was sort of accidental in that, with the best will in the world, if I'd gone to Ogilvy, they had said—I think there's a real opportunity—directly, they would have thought, basically, "You're a fat, 59-year-old man. You know, I can't think of anybody less suiting to the medium than you, to be absolutely honest." Okay. And you're talking about marketing stuff. What happened was a very talented young guy, aspiring filmmaker. Okay. After—under COVID—under lockdown, I did a hell of a lot of podcasts. I'm already up there on things like TED Talks and a few other things, which gave me initial kind of recognition. And during COVID, I thought, well, I can either talk to 250 people or a thousand people on a podcast, or I talk to nobody, so I might as well talk to 2,000 people. And they're pretty much the simple maths. And so there's a hell of a lot of footage. By the time you got to about 2003, there was a lot of footage. And this aspiring filmmaker guy just started taking it and effectively TikTok-izing it. I.e., you make portraits, you bang subtitles on because the kids are watching in school, so they have to have the sound turned down, and you put in some nice animations. He was doing it brilliantly well. At first, I knew was basically when I started getting kind of mobbed by school kids and my children getting kind of half happy because their dad's famous and half alarmed because your dad isn't supposed to be on TikTok. It was very—my, my children had this completely bifurcated reaction. But the point was, I think that a large part of marketing, if we're being honest about it, is simply increasing the—the—scene talent would say—you increase your surface area exposure to positive upside optionality. In other words, you know, if, if you are more famous, if you are better known, if you're a better-established brand, you simply have more ways in which you could get lucky. Can you predict those in advance? Could Guinness have predicted splitting the G? That promo—I mean, maybe there's this genius secret cell, you know, within Diageo who are manipulating us all—and—and—you—but actually, you—some other part of marketing is just spotting what you might call the meteorology of human behavior—spotting a trend, you know, an area of high pressure as it were, and responding to it, you know, in other words, "Well, this thing has fallen into our lap. Is it right that Guinness has moved the G up on their glassware slightly? I think—to, to make the first drink—the first gulp—slightly smaller." So they're responding to it—that their antenna quite rightly—and that's, you know—I mean, a large part of marketing is exactly that. It's just having your antenna sufficiently attuned to what's happening that you're not left looking stupid or that you take advantage of things when they, they fall in your lap. And I, I think the problem—okay—finance—when it's bad—and according to Roger Martin, he thinks that three-quarters of CFOs have a net negative effect on their business, and he thinks it's only the top quartile who are real value-add guys. Okay. And the two problems you can have from bad finance is one—an obsession with certainty. Okay. So, in other words, you'd rather have a definite 3% than um a possible 15 to 20%, even though the odds are far better of the latter. So an obsession with absolute certainty and predictability, which is—basic—and also the instantaneousness of results. Now, if you demand complete certainty and instantaneous results, nine times out of 10 you're going to resort to cost-cutting. Right? The second thing that Roger Martin—who's speaking at Nunchtock, by the way—do you know this? He's—I think he's the heir to Peter Drucker. I think he's fantastic—brilliant, brilliant business strategist—uh, who's basically a marketer by temperament. Okay. He believes that competition takes place on the shelf and in the mind, not in the factory. Okay. I write—okay—the second problem with finance is demanding ridiculous levels of proof before you can do anything. And again, of—okay, that will always bias you towards cost-cutting rather than investment, but it also makes any kind of imaginative marketing almost impossible because, as Roger says, there's never any data about the future. And so I think, you know, that problem—effectively—where we, we made this mistake as marketers—going, "Yeah, we can be—we can be just like operations—we can be absolutely quantifiable—everything behaves just like Newtonian physics." Okay. And I think that was a—I think that we—I understand exactly why people wanted to say that, but only a small percentage of marketing activity falls into that category. Tends to be bottom of the funnel, and that—and actually only for a small percentage of businesses—fast-feedback businesses—can kind of test and learn at the bottom of the funnel very quickly. Okay. In B2B, I mean, you know, forget it. And therefore what we've done is, by making this claim, we actually painted ourselves into a corner, which is—instead of saying, "Hey, marketing is great because it's quantifiable"—we found ourselves in a place where the only marketing we were allowed to do was quantifiable stuff, and we should have been making the case—"Look, a large sway of this is basically—it's—it's a casino. Okay. Think of it—blackjack—not roulette. Right? It's a casino where skill matters, but where you can't card count. And weirdly, the odds are stacked in your favor." As Jeremy Bullmore said, the number of cases where someone does an ad campaign and the effects are actually catastrophic—it's actually quite rare. Okay. So it's a—it's just a great casino. But we shouldn't pretend that we—we—we can tell you to the penny how much cash you're going to walk out with.

So I, I, I want to come back to the thought and the observation then um, around your TikTok fame—is the fact that for many, many years before you were constantly—let's call it—being in the game—in the sense of you—you were in—you know, you were doing those TED Talks—you made an active choice therefore to do all those podcasts for all that time—for those—there's a very funny reason for this, by the way, which is if you do public speaking or podcasts, you have to do it fairly frequently because otherwise it's too frightening. It's a very weird psychological—

Thing. So a lot of people I know who are on the speaking circuit, um, you tend not to do many speaking engagements in July and August. Okay. Uh, the really hardcore public speaking people actually go to Australia or Singapore for July and August and pick up the slack in the other hemisphere. Right. But what we all noticed is that when we have to give our first talk in, on sort of September the 7th, we're bricking it. Okay. So there is a weird psychological phenomenon, which is why people either don't do public speaking or they do a lot of it; which is, if you do it in—it's a bit like the reason I always say one of the reasons Londoners never move out of London—okay—is if you only drive infrequently, you don't enjoy driving. Okay. It's a really—so first of all, driving in London isn't much fun anyway, but then these guys either don't own a car, they rent one every 3 months. Now, even I, who drive a lot, when I get into a rental car, I don't enjoy the first half hour because everything's unfamiliar. I don't know where the indicator stalk is, and then about 45 minutes in you start to actually get into it, become a system one. Okay. And it suddenly occurred to me that public speaking is like driving in London. It's like driving. If you don't do it a lot, you actually think of it as a horrible thing to have to do. Once you move out to Seven Oaks, that, you know, I still get a sense of joy, uh, leaving the house and hitting 40 miles an hour within, you know, because having lived in London, this is still, you know, a wonderful source of, uh, of happiness. So, you're so you're constantly rubbing the muscle. You're kind of using the muscle constantly, which gives you that that sense of it.

And and now today, clearly, you know, whether we call you a celebrity or an influencer, the one thing that's undeniable is the fact that you've been able to carve out quite a unique personal brand. In my defense, there is a strategy behind it. Okay, which is, if you want me to be completely candid, I think advertising agencies within which I work will have spent all their time trying to impress the marketing director. And my argument is unless marketing is more widely respected within the organization and within society, um, then actually trying to impress the marketing director is actually a misdirection of effort. Yeah. Okay. That actually our problem is a category problem; it's not an individual agency problem. What I then discovered—totally my accent—is that the appetite for consumer insight and marketing insight outside what we'd think of as the core target audience of marketing people in large organizations, the appetite for this stuff is almost limitless. The other discovery I made, which is really fascinating, and that's by the way one of the reasons why you market a bit indiscriminately, is that it prevents you—become—it prevents everything becoming a self-fulfilling prophecy. Okay, you—the other thing I discovered was a very interesting psychological thing, which is whenever you talk to an entrepreneur and you tell them a counterintuitive insight about human behavior, they're all over it. They're absolutely fascinated. Okay. If you talk to a bureaucrat, they get angry. And I've come to the conclusion there are fundamentally two mindsets. There are people who want to have a neat artificial model of the world. And information that conflicts with their pre-existing model, they actually find annoying or or they desperately try to dismiss it. Then there's the entrepreneurial mindset, which sees any unusual behavior, however anecdotal the information may be to support it, as an arbitrage opportunity. Okay. Okay. So I can actually be—be perceived as much better by my customers without spending a lot of money if I focus on this particular area of perception rather than that. And they see it as basically it's a competitive source of differentiation. And of course, the the other mindset, which is technically called the Ricardian vice in economics, but I won't go into this, which wants to adhere to the model and basically forces the territory to conform to the map—that mindset is completely opposite. So the first entrepreneurial mindset views dissenting or unusual or contradictory information as basically an opportunity, and the other group fundamentally see it as a threat, which is why so much of my time was spent talking about trains.

Yes, because you realize that transport economics is entirely in the grip of this speed, capacity, time fit. And I came along and goes, "Hold on. I had this this morning, right? I had to walk along three carriages of my train to find a place where I could use a laptop where there was a table." Now, who is it who thinks that it's okay to spend 120 billion reducing someone's train time on a trip to Manchester? But it's not okay to spend 400, you know, I don't know, £30,000 putting tables on trains. Now, literally, I—if you've had tables on all the trains between Oxford and Cherry Cross, okay, that would mean that would mean I could work for an extra—um—I don't know, 200 minutes a week, right? I don't get to Manchester all that often. Okay? Now what I'm saying is that trade economics was patently—in order to reach this level of wonderful mathematical—Leturnian certitude—they basically froze psychology and behavior entirely out of the mix. Now it's worse than that. Okay. So there's Dan Davis—I recommend to everybody his book. I like to keep it Welsh as you know. Okay. Right. Okay. Um, uh, Dan Davis, uh, argues that the reason Boeing, uh, got—trout—by I think it was McDonald Douglas, was it who they bought—right—but the McDonald Douglas culture trans—the Boeing culture, and his theory is that the McDonald Douglas people always had numerical arguments and the Boeing people could only respond with things about culture, and you have this fundamental imbalance I think, which is a kind of quantification bias, which is the person with data—even though it's an aggregate and an average and it all comes from the past—always beats the person with abstract nouns. So Roy, just to pick up on this point about the bifocation of mindsets, remember Dr. Helen Taylor, a sort of mutual contact. Oh yeah, this is killer. And um, she talks about the exploration exploitation construct, and and I think you refer to it in the context of—be—by the way, guys. I mean, I would argue that that the ex—the explore exploit trade-off, which isn't a trade-off, it's actually it's a Mobius strip. They're two—if you were—if you had an Asian mindset rather than the western mindset, you'd see it as yin and yang, is in other words, the union of opposites—complementarity—and actually the exploitation funds the exploration, but the exploration informs the exploitation and keeps it current and contemporary and relevant, which is very much Helen's argument is about comp—and that's why we have neurodiversity by the way in humans—yeah—so so linking it all the way back into society and education, how do—how do you create a context for more exploration—one of which is—I—you need fundamentally different, uh, accounting principles for innovation and marketing, which are—they're not exclusively—marketing is not exclusively explore by the way, that would be—nor is innovation by the way. Okay. Uh, but the waiting towards exploration of what you do—of what you don't know—is higher than it would be in operations or logistics. And by the way, there should be explore in operations and logistics. I'm not—I'm simply saying that we will have a greater proportion of scout bees in a marketing function than you might want to have in some other, uh, internal—in purely internal function—where psychology doesn't feature and where, uh, the opposite of a right answer is wrong. Okay, we're in a very—we're in a much more ambiguous space because sometimes the right answer is to do two opposite things, for example. Yeah. And you know, often in a in a kind of Newtonian environment, the average is actually a useful indicator. I would argue in marketing the average is a dangerous distraction because the act of averaging removes the outliers, which—which is where the real—the real insight can be derived. Okay. So you you simply need to accept that—not for all of marketing all of the time—but you fundamentally cannot apply a simple co—you can apply a simple cost-benefit analysis to the exploit bees. Okay, it's energy consumed has to be less than energy recovered. You can see why accountants like that because it's a double-entry bookkeeping model and it's real time pretty much. Okay. The explore bees, you have to accept the fact that some of their journeys are a waste of time. And it's interesting because a lot of people don't like Musk or Bezos—that, you know, they don't like those weird people—or Zuckerberg. But certainly Bezos and Musk, uh, both have an interesting decision-making style which allows for more explore. Um, notably, by the way, everybody except, uh, Bezos hated the idea of Amazon Prime. He pet-forced through it. Bezos has a really important concept which I think we ought to acknowledge in marketing, which is the two-way door. Type one decisions, type two decisions. Okay? And a type one decision is one that—one spade—is irreversible. So you're opening a 20 billion square foot warehouse north of Nashville, right? You—you can't really double back on that, or you try, but I mean, okay, it's—once you've done it, you've done it. Okay. And Bezos argues that we demand the same level of rigor for type two decisions, which are reversible, as we do for type one decisions, which are not. And he thinks that's a fundamental mistake. And the interesting thing was that Amazon Web Services, the most pro—AWS, the most profitable bit of Amazon emerged because the original business case apparently wasn't all that good, and people were pulling it to pieces, and Jeff just goes, "This is a type two decision. It's a two-way door. We've got to invest in all this infrastructure, bandwidth, and uh, compute anyway because we're Amazon. If we can sell that to other people, we'll be doing fine and dandy. And if we car—what have we lost? We've leed a bit of a lesson." Okay. And I think there's absolute case, which is it's also compounded by the fact that it's business. It's very easy to make ass-covering look like rigor. Okay. In other words, you appear to be a very, very rigorous decision maker who's demanding to know absolutely everything. Okay? You'll never get into trouble for looking rigorous, but what you're actually motivated by is ass-covering. I don't want to get blamed if anything goes wrong. And so I think—I think we just need to accept, uh, and and actually Roger L. Martin, uh, who's—as I said—my Svengali in all marketing matters. He was the dean of the Rotman School in—um—at—in Toronto, extraordinary guy, transformed—among other things. He worked with AG Lafley of P&G for many years. Transformed Canadian tennis. Extraordinary guy. He makes the point that the, uh, fundamental thing that distinguishes a good CEO from a bad CEO is that a good CEO acknowledges that business is probabilistic. It's not deterministic. And often when a CFO becomes a CEO, of course, uh, it's very uncomfortable for them to move from a deterministic world to a probabilistic world. Uh, so I think—I think that you simply have to accept in marketing—look—um—I—the way I describe this is that in certain areas of business you can be right, okay, and you can use rationality, and rationality is a bit dangerous because implicit in a rational argument is not that just the fact that you're right, but the implication—the connotation—that everybody else is wrong. Now, in marketing, the best you can do is probably not to say this is absolutely the right answer and anybody who does anything else is a fool. What you can probably say is this is worth trying. This is worth exploring. Okay. And what you can consequently say is maybe the job here is not to be right. That's too high a bar. What we need to be is interestingly less wrong. And actually when I say interestingly, interestingly implies in a way different from our competitors, or that's unusual, or that's, uh, that attracts attention, or that's noticeable, or that intrigues people. And that's probably the best you can do in an ad. Be interestingly less wrong.

Rory, I'm going to come on to talk about cars and I've got to—I've got to ask you about perhaps one of the most controversial campaigns of last year—Jaguar. What's—what is your take on—on genius or or fool? It might be genius. Let me explain. Go. Okay. The first thing we got to remember is that when something like electrification happens, it's an iPod moment. Right now, I'm totally in awe of and totally defer to people like Mark Ritson, you know, Byron Sharp. They're basically right about brands in a stable—in a relatively stable state environment. Okay. And and by the way, brands that are already quite successful, right? If you're, you know, if you're the brand manager on, you know, Tide, okay, in the US, take an extreme case which has like 70% market share, okay? You basically worship at the altar of Ritson and Sharp all the time and you won't go far. They're basically right. Okay, there are—I think—except—well, obviously there's an exception which will be, you know, supermarket-owned brands don't conform to the Ehrenberg, you know, principles because, you know, which own brand you buy depends on where you shop—I mean, okay, so there are exceptions like that—I don't think—I don't think those rules are much good necessarily for disruptive brands that are in a hurry; they might be okay for disruptive brands which are really, really patient and have a lot of money to spend, but—and secondly, I don't know if they tell you what to do when effectively there's a seismic shift in your marketplace. Okay. So, the point I'm making about Jaguar was that there, okay, there are a lot of things. Okay. I really like red post boxes, right? Gilbert Scott red post boxes. I don't think they should be the BT logo or the or the EE logo, right? You know, it's also rearing that there are a hell of a lot of people who absolutely love Jaguars and the Jaguar brand, but they don't actually buy them. Sometimes, and this is a very, very—surrounded by caveats—statement, okay? Be really careful about this. Sometimes you've got to piss off your existing customers and find new customers somewhere else because you can only profitably make electric cars in the UK, uh, and compete with Chinese incomers if you go for the super premium segment. And one really important thing about Jaguars, I think, is they're built in the UK. And building things in the UK is expensive. Okay. So, you have to go for a level of customer—globally too. Okay. Which may not be the people who have affectionate memories of, um, driving gloves with, uh, you know, web backs and Lucy Clayton girls being taught to get out of an E-type without showing their pants and the XK120 and Brooklands. Okay. Right. So, sometimes to preserve the business you do have to—bo. Right. What—what happened? He went electric. Someone shouted Judas. Okay, this is exactly the same. You have a folk brand that needs to become an electric brand and people are shouting Judas, right? Okay. And there are just cases where—I mean Coldplay famously, right? Okay. Basically alienating a whole audience. The Beatles and and and by the way, Rick Rubin—really good. I did a Spectator interview with Rick Rubin. He said basically there are two—there are two kinds of musicians. There are musicians who always reinvent themselves in the kind of Eno way where they never step still. Bowie, you know, Eno, you know, the Beatles to a large extent. And then there are kind of musicians who just basically just plow the same furrow really, really well, never change. And sometimes they'll lose an audience because of fashion—Johnny Cash, right?—and then win it back because—act—we were talking about this yesterday actually. Sometimes with a brand, you just keep on going. You fall out of fashion and you just wait. You know, as long as you're keeping yourself alive actually, you know, the market will change, you know, things will change, what's cool will change. Um, you know, I—I probably wouldn't do that with a shell suit. You know, I—if I were a shell suit manufacturer, I'd kind of go—that ship—as we were talking about this, by the way. It never made it to London, did it? There was that weird period in like the 90s where you got on a train up north and about an hour out everybody was shiny. Do you remember that? And we were trying to work out where it came from and why it never made it to London. But—but actually sometimes I would defend the fact that Jaguar didn't change their name. I mean, okay, they got—they didn't get rid of the logo of the leaping cat either, um, the design of the car is absolutely correctly—as it should be—an extremely innovative design because that's why Jaguar exists, and then these people started writing about the line—a copy of absolutely nothing—the lionists of William Lyons in like 1928—he said it right—he wanted the Jaguar to be a new car—the XK120, the E-type, all those Jaguars were freakishly innovative and extraordinary cars when they first appeared; they weren't derivative. So my argument is—look—I—I love the Jaguar brand. I mean, I've—I've had seven Jags I think over—never too—sort of—about—I rent a lot of Jags. Probably can't afford this new one, but I'll still want it really, really badly. Okay. But I want the company to survive, the business to survive. And I don't think you can be—I don't think you should be a marketer without actually having enough respect for the financials of a business. That—that is one thing I think—by the way—where, you know, I'll criticize marketers, which is sometimes you'll talk to a client and you go—let's say—I'll give an example. You're talking to a hotel client, okay, and you're interested in, say, yield management or whether you could sell to people late at night because, you know, you've got a load of rooms free, and you go, well, what's the cost of turning over a room relative to having a room empty? Okay. And the client doesn't know, and you kind of go, you can't actually do your job if you don't know that stuff, right? You know, because you can't be having really imaginative ideas for who could be occupying an unoccupied room and at what price if you don't know the cost of cleaning a room versus just leaving the room empty. And so, one of the points which I think we ought to make is we shouldn't criticize Jaguar's marketing strategy until we know what was the data about the future of the car industry because, okay, they could end up—okay, I don't want—I—I love Jaguars. I love all the—about leather over bonnets and, you know, uh, you know, you know wonderful speedometer dials, Smiths, etc. I love all that heritage stuff equally. I—I want the business to survive. Okay. And I—I, um, you know, I acknowledge the fact that I, you know, it's a brand I'm nostalgic about. But BlackBerry—Black—and tragically BlackBerry was kind of right. You type much faster through a keyboard than you can on a screen. Okay. But fundamentally there are moments where the entire c—I mean, you're in the insurance business, right?—I mean, literally when my father died, I discovered there was something called an insurance broker—going—what the hell is all that about—right—but there was a time—and that's how you b—insurance—so—so, um, so the jury's out on Jaguar, but I heard an expression—oh yeah—I heard something the other day which was that the quote—jack of all trades, master of none—Mhm—is actually—jack of all trades and master of none is better than master of one. I never knew that. So all this time society has been getting that diametrically the wrong way round. I like that phrase T-shaped, which is you have one area in which you're deep and another dimension in which you're broad. And I—by the way—in marketing the great thing about marketing is if you—it's basically a great way to monetize ADHD—cate—because, uh, you know, if—if you're in logistics then sitting in a Paris cafe or watching your beagle on a train doesn't make you better at your job, whereas in marketing it does. But where is a brand or marketing trotting along thinking one thing and getting it diametrically wrong? Right. Okay. One area would be not a brand but a category, which is car electrification. Okay. There are a lot of—I would love a world—because obviously, you know, I'd be better paid than—Okay—where the way consumer choice goes is—okay—um, uh, category—brand—channel. Do you see what I mean? Okay. Sometimes the order of the choice—decision-making tree—and it's a practice known as elimination by—Okay. So when you buy a house, okay, actually Rightmove forces you to buy a house in a stupid way in my opinion. Okay, because you basically go—house or flat. Well, I probably know—I want a house, but a really good flat is better than a house, right? Okay—house or flat. Then it says, okay, you know, buy or rent. Then it says, where do you want to go look? What price do you want to look at? Now that assumes I know exactly what I want in advance, and people don't. But it also drives you down a particular order of decision making. Now, real-world human decision-making is actually iterative. So we use what we find in search to refine our preferences. So our utility function changes according to what we find. And so with David Ogle, who famously crossed the Atlantic to buy a house in Paris and ended up being sold a chateau 600 miles away. Okay. We don't know what we want. Actually the estate agent knew what David wanted better than he did, I suspect. Okay. And so what's really important is that I think there are a lot of cases where people are—and Dave Trott's very good on this. He says if you're a—if you're a challenger brand, you focus on your—you—you advertise the brand. If you're a brand leader, you promote the category benefit. Okay. I think there are a lot of cases where we're advertising one brand at a time when actually what we need is category advertising. And I don't think it exists anymore, uh, much. It used to—in the 1930s the way it worked is you had—man—campaign—if you want to get ahead, get a hat—right—and the way it worked was the Hatmakers Federation of Great Britain or whoever the sort of trade body was would levy like one or two percent on sales of all their members and they'd use it to fund an ad campaign encouraging people to wear hats—not one brand of hat or another brand of hat—but hats in general—and I think the reason it doesn't work is globalization, which is then, you know, 90% of incremental hat sales would probably be made in Britain by members of the British Hat Federation, so they could afford to advertise the category—in the case of—um—in the case of—Got Milk is the example—okay—it's the Milk Marketing Board—it's advertising milk as a category, not one brand of milk over another brand—clip—in the case of electric cars—that—when you should do category advertising—ing—and when you should do brand advertising. Ad agencies basically are like bloody pandas, right? They only eat one thing, which is a brand comms budget. This pisses me off massively, right? Because not all problems are a brand. Okay, there is scope for creativity right across all the Ps of—however many Ps of marketing you want to define. There's scope for creativity in pricing. There's scope for creativity in product. There's scope for creativity in channel, for example. Okay, which doesn't begin with a P. Um, now, ad agency people are obsessed with this idea that you only solve the problem if you've defined the problem as a brand comms product. But I would argue—depending on how consumers make a decision—there are decisions where brand is absolutely primary. Okay, you're standing at a shelf and there's Persil and there's Ariel, right? That's—okay—a brand led with price and a few other things at facings. I mean, there's always other going on, right? But yeah, that's a brand decision, right? But whether you drink tea or coffee is a category decision. Whether you buy an electric car or a petrol car is a category decision. Now what's happening in electric cars? Everybody's going—our range is a bit better than their range. What you're doing is you're making the element of comparison between electric cars—range—which is the only negative of the category. Okay? It's a bit like if you imagine tea drinkers wanted more people to drink tea and less people to drink coffee and every ad said—our tea tastes a bit less than their tea. Yes. Right. Okay. You would get many people drinking tea? No. I think someone—government—I mean a new version of the COI—if we're going—

To spend literally billions on the electrification of, you know, the upgrading of the grid, of green power networks, you know. I mean, great Jackson; that octopus is bang on the money, you know. He, he's the person to speak to, okay? For you spend billions of this, we spend billions on battery tech; shouldn't we spend a few million quid just telling people 11 ways in which electric cars are better than petrol engine cars? And I don't mean environmentally, okay? I mean simply from the point of view of a selfish motorist who likes driving, okay?

Um, let's put it really simply, okay? I mean, uh, a typical uh internal combustion engine car probably contains about 207 moving parts. Lots of things need replacing: oil filters, air filters, all that stuff. All of those things exist to perform a simple function, which is to rotate a shaft to give you forward momentum. Right? Now, Michael Faraday came up with a thing. It's a bit debatable; it's probably Michael Faraday, okay? Came up with it called an electric motor, where you put electricity in, the shaft goes round. Okay. Sorry. Right. Okay. Anybody with any clue about engineering, any alien who'd come down and looked at the internal combustion engine would go, "It's Heath Robinson. It's brilliant." The fact that loads of Germans have managed to make it reliable, quiet, and smooth is fantastic. But there's a saying in the car industry: the best part is no part. Okay.

I'm really worried. Right. This is why I'm worried. Okay. Because the problem we have in Britain, the biggest marketing problem in Britain, okay, isn't like whether people buy PEL or Ariel. The biggest problem in Britain is the Daily Mail. And I don't mean the Daily Mail specifically. If I, I always said to Martin Sorrel, what I would have done if I'd been Martin Sorrel is I just would have gone to the British newspapers and said, "Right, we're going to pull all our ad spend if you don't cheer up." Because the negativity bias in press, particularly towards innovations, right? So you have electric cars, which are basically a great car news story in cars. What's going to happen in about six months' time is there's going to be a massive number of people who leased, took a three- or four-year lease out of an electric car, and those cars are going to be hitting the secondhand lot, which means that both of you have, you got electric already, you have hybrid, okay? That's sort of the bisexuals of the automotive world, okay? Hedging your bets, yeah. Okay, they're going to be young people can go and buy a three- or four-year-old electric car, which is nearly as good as a new car, particularly to the extent that the software has been upgraded over time. Okay, for bugger all money. That is a fantastic news story. What will the Daily Mail say? It will say car forecourts are clogged with unsold, unwanted electric cars. Okay, so instead of get a far better car than you could have dreamed of owning five years ago for half the price, they're going to be saying, "Oh, nobody wanted all this stuff."

I got a friend, a guy called Bobby Healey in in Dublin, right? Runs a company called MANA, which is drone delivery. Now, I'm going to confess something here. When I first met him, and whenever anybody's presenting drone delivery to me, I've gone, it's technology and searchful application; it's a load of bollocks, but okay, but you know, it's cute, but you know, let's face it, you're a load of people who like drones who are just looking for use cases. And then he explained the economics, which is he takes over like eight parking spaces on the roof of a, it's, it's suburban Dublin is the main area for delivery. Takes over eight parking spaces on the roof of a shopping center in Dublin. Downstairs you've got access to coffee shops, clothes shops, chemist shops, all the people might want to deliver. Okay. Someone places an order. Someone comes up from the shop, pops it in the drone. It then flies off at about 70 miles an hour, covering about two miles in about two minutes, right? It lowers the thing onto their designated landing spot, which in 20% of cases is the trampoline of their garden, which has been repurposed as a helicopter LZ, right? And then it flies back. The cost of that delivery is about eight pence. And it was when he told me that, but his problem is entirely psychological. Was not tab, which is okay, during the time he's been operating something like, you know, I don't know, 10,000 moped journeys have been taken off the road, reducing congestion, reducing emissions, but also three people haven't been hurt in moped accidents because the drone's been delivering, right? But that's invisible because we don't see the accidents that don't happen, and the Daily Mail isn't interested in reporting a moped accident, right? Or the absence of a moped; that's even less of a story, right? What's the worst story you have in the time of newspaper? Nothing much happened. Okay.

By contrast, as Bobby knows, all it takes is one can of whiskers to like injure someone's pug, okay? And it'll be on the front page of the newspapers. Now, I don't think I grew up in an age, you did, where we were actually optimistic about possibilities. And I think we have a media environment which is so biased towards negativity that things we should be excited by, we should be making heroes of British companies like Octopus, AOD, right? Uh, AO, they're magnificent billion-pound British success stories. Uh, you know, Monzo, Wise, okay, Direct Line, EasyJet, okay? All of those companies which are basically, by the way, marketing innovations. Let's be, let's be honest here, right? Basically marketing companies that they're companies that exist; dice. Someone spotted a psychological um uh what, wormhole, you know, in the human brain. Basically, that's why those companies exist. You actually, weirdly, if you put your phone number on the front of, on the side of a plane, nobody goes to travel anymore; they come straight to you, you know, all that stuff. Okay? And instead, we in America still, you know, despite all their wackiness, they still make heroes of people who do significantly innovative things. And he said we basically look, you know, look for a reason for them to fail. And with electric cars, a marketing failure, I mean, one of the things I said, which is quite interesting when you think about it, which is people go, "That's expensive." I go, "Well, compared to what?" Well, they're just expensive. I go, "No, no, an electric Skoda is more expensive than a petrol Skoda, but in terms of driving dynamics and experience, a petrol, an electric Skoda is closer to a petrol Audi." I, I might even argue that if you take away the really expensive kind of suspension, in some ways the driving experience is closer to a petrol Bentley than it is to a petrol Skoda. Okay, you don't get the leather, you know, I'm not claiming, but in terms of a car that can be, it's quite difficult in a petrol car to have one that's both a limo and a go-kart, and you can do it, but it's Aston Martin Bentley territory, right? You've got to have a massive engine and you know, and then it weighs a hell of a lot. Actually, nearly all electric cars, you can drive them like a limo or you can drive them like a go-kart, you know.

And the interesting thing, I'll give you the exact analogy to this. This is why this is a particular case where marketing needs to be a special case. Okay? It's where it's what I call Japanese toilet cases. And someone actually, Roger Martin again, sorry to plug him all the time, uh, he's actually written a piece about this on his Medium post. There are things where people are resistant, but once you get them to try the new way, they never go back. So that's mattress toppers, it's air fryers, it's um, the radio automotive example is automotive transmission, automatic transmission, sorry. Right. Okay. Everybody in Britain, you grew up going, "Oh, I wouldn't like an automatic. No, they're no good. My grandfather had one; they're rubbish." Okay. Everybody resists, oh, I lose the sense of control. Every single sodding person who who buys an automatic never goes back, right? And I think electric cars are like that; if you look at the data, everybody really, really resists them. And what we need to do is find out a way to sell people. Actually, the way to sell electric cars is really easy: it's a two-week test drive. Okay? You, once you've got a two-week test drive. Don't make, this is behavior, don't make people hand over their petrol car on the day they take irreversible delivery of an electric car because it'll frighten the hell out of them.

Similarly, environmentalism, right? Let's just look at the degree of difficulty involved in getting people to a heat pump, right? What they want you to do is rip out your gas boiler, okay? Which has given me trusty service for 20 years, and dig up your lawn and put in a heat pump in your house. Okay. But I go, those are about three things which are behaviorally really, really difficult. In addition, you have to ask people to understand the physics of heat pumps, which are deeply counterintuitive. Yeah. Now my brother's an astrophysicist. He's doing his environmental duty by basically just one room at a time. Not all at once. Right. One room at a time, sometimes two rooms at a time. He's installing air conditioning in his house. Right. Now wait for it. What almost nobody knows is an air conditioning unit is an air-to-air heat pump. You just make it work backwards, right? They've been sold for ages; they're made by well-known brands like Panasonic that we've all heard of. Installation's really easy. You can still use your gas boiler on the five days of the year when it's really solid cold. But the rest of the time when it's this temperature, which is what Britain mostly is, it's like 46, right? Oh, sorry. Fahrenheit. Last person. What's that? What's that in the new money? As that's seven, seven or whatever. Okay. When it's this kind of temperature, okay, a heat pump, air conditioning unit working backwards will do a fantastic job warming your home. Your boiler won't have to do a thing, right? And it's sort of 5 to 1 efficiency. So I said to my brother, "Oh, presume you get a subsidy for that?" No. Said, "You only get a subsidy for an air-to-air heat pump if it isn't an air conditioning unit." I said, "Well, why is that?" He said, "Because they're frightened people will use air conditioning to cool their homes and it'll increase energy consumption." To which I said, "Mate, this is the UK, right? That is a really, really good government policy in like Morocco or Greece, right? It's a really good policy in Arizona. In Britain, two things. One, yes, my brother uses it in an air conditioning unit on perhaps seven nights of the year. You know those nights in the UK, you get about seven of the year where you can't get to sleep, right?" Okay. Secondly, it is good because old people die on it when you get a heatwave in the UK. So allowing old people to have a place they can go which is cool will save a lot of lives. Thirdly, the amount of it when this happens, solar, solar energy is probably being generated in huge quantities. So you don't have to worry too much about the, but these people, what, what marketers understand is that the perfect is the enemy of the good. You have to make behavioral compromises. Can I ask Anna say or any of you to go look, it's too grand to do a room and you get an air conditioning unit, but you also get a thing that heats the room most of the time really effectively, and you know, you can use it on clever tariffs, and I, I, I can imagine saying that, right? Can I get someone to say, rip out your boiler, replace all your radiators, dig up your lawn, right? So the fact that no one consults marketers in terms of what's an easy sell and what's an impossible sell when they design government policy and they simply use economic incentives as the early tool. That's where this is where total idiocy prevails. Thank you. A fascinating, fascinating conversation.