Roryisms register

Rory's most quoted line is, he confesses, really Niels Bohr's, said to Einstein: in simple physics the opposite of a good idea is wrong, but in high-complexity physics, and in psychology, it can be another good idea. Every algorithm assumes faster is better, yet there are two great ways to check into a hotel: fully automated, unlocking your room with your phone, or someone taking you up and making you tea. Guinness turned a pour so slow that bar staff hated it into 'good things come to those who wait'. Retail trends towards self-service and low prices; the farmers' market does the opposite, a Tesco Metro made much worse and dearer, and people love it. So look at your biggest competitor and do the opposite.

“In the real world, the opposite of a good idea can be another good idea.”

The point: Human value is not monotonic, so optimising in one direction freezes out equally valid opposite solutions, and the middle often loses to both extremes.

Variants: Examples rotate across 19 tellings: Dyson versus Henry, no-touch versus high-touch hotels, a hamper turning a slow Waterloo train into a day out, bargain and luxury both winning while the middle suffers. Sex, he notes, is never improved by being high-speed.

In business it is dangerous to be slightly bonkers: try something odd that fails and your job is on the line; try something rational that fails and you get another go. That bias means you gain an extraordinary advantage by creating a small space to test things that do not make sense. Competitors are too scared, or culturally incapable, of copying, so the advantage lasts. Just as bees let some of the colony fly at random, ignoring the waggle dance, organisations should permit experiments outside the rationalists' comfort zone. A performance marketer found that people who tested eight things were disproportionately more successful than those who tested four: with eight, you make room for a couple of silly ones, a green aardvark, not just price.

“Be weirder.”

The point: Deliberately testing illogical ideas is where outsized breakthroughs come from, because rivals filter them out and breakthroughs are disproportionate precisely because they are illogical.

Variants: In a 70/20/10 budget, aim for a high failure rate in the experimental 10 per cent or you are not being brave enough. Another telling adds that behavioural science sometimes works even when wrong, simply by licensing tests a rational person would never run, as with Fleming's accidental penicillin.

Find a body and police go door to door asking an open question: did you notice anything unusual last night? Someone says a white van drove past three times. That has no evidential value; you cannot jail someone for driving a van oddly. But it tells you where to look next: check the CCTV, and if the driver turns out to be a known weirdo, reweight your probabilities. Police work in two phases, investigative then evidential, because they deal with imperfect information and witnesses who lie. In their approach to information, Rory says, cops are cleverer than lawyers, and would have handled Covid better than scientists. Business, aping economics and physics, has skipped the first phase and demands evidence before allowing any speculation.

“It has no evidential value, but what it does tell you is what to investigate next.”

The point: Separate a speculative search phase, where weak signals and anecdotes direct attention, from a proof phase; demanding evidence too early blinds the search.

Variants: The van passes three, four or six times. Some tellings add that police act to create evidence (a stakeout, a honey trap: 'do to know rather than know to do'), and that science, like a trial, later writes up messy discoveries as if planned, as Peter Medawar complained.

Rory hates owning anything that isn't dishwasher-proof, so he proposed a solution to his wife: treat everything in the kitchen as if it were. After a couple of years, whatever has survived is by definition dishwasher-proof. Kitchen-utensil Darwinism. The same filter has been applied to the world's problems. There is no shortage of rational people, McKinsey, finance departments, procurement, all trying to look logical to each other, so the problems logic can solve have been solved. The ones that persist are probably logic-proof: most political, business and, he suspects, marital problems. If there were a logical answer, we would have found it. So for an enduring problem, try something else: intuition, conjecture, randomness, comedy or luck. Logic has its place; he does not want a conceptual artist running air-traffic control.

“If there were a logical answer, we would already have found it.”

The point: Problems that have survived years of rational attack are filtered for logic-proofness, so they are exactly where non-logical approaches are most worth trying.

Variants: The survival period ranges from three months to two years. In one telling he uses it to defend unconventional political moves (the rational people had 20 years and failed); in another it is his eleventh rule of ten.

Starbucks is great if you have 45 minutes, but wrong at a station, where you want to grab a coffee and catch the 6.47 without queuing behind someone ordering an elaborate iced drink with whipped cream. So Rory's coffee chain sells one thing: flat whites, made in advance, with oat milk the one concession. Tap your card, pay a fixed price, pick one up and go. It is the Henry Ford of coffee. Ask for a cappuccino and the staff point to the logo: the answer is in the name. Others have borrowed the idea, and after some publicity conference organisers called. Everyone pours out of the plenary at once wanting a coffee and gets bad filter coffee or a huge queue. A market the inventors never imagined.

“If you ask for anything other than the flat white, the answer is in the name.”

The point: When a whole market chases customisation, a radically restricted offer can serve an overlooked need if the trade-off is made explicit, and fame then brings uses you never envisaged.

Variants: One telling casts it as Rory's retirement plan for railway stations; another stresses that others spotted the conference market after publicity, an Adam Smith division-of-attention point.

A Hindu friend of Rory's (Jag Bhalla, in one telling) accuses Westerners of 'monotheorism': just as we insist on one God, we insist on one theory to explain everything, or one metric to decide everything. His mother goes to a temple in India whose altar holds an elephant, a monkey and Jesus, and she sees no contradiction at all; they are parts of the same thing. The Western mind treats anything with two modes of behaviour as a trade-off or a conflict, and has built a faux rationalism in which there must be one right way. Rory's view is that complex systems need several partial theories held at once, and that a great business must cope with pluralism: doing what it knows works while also exploring the unknown.

“The trouble with you Westerners is you have to have one theory to explain everything, or one metric to decide everything.”

The point: Insisting on a single theory or optimand turns complementary approaches into false trade-offs; holding several partial models at once is often wiser.

Variants: Applied in turn to economics (Rory prefers its fringes, Georgism and the Austrian school), to marketing (The Long and the Short of It is both, not either), and to business pluralism, with Buffett's 'if the phone don't ring, you'll know it's me'.

In the 1980s and early 90s the PC industry's nerds competed on processing power, RAM and clock speed, and neglected how computers looked or felt to use. Put a beige or grey IBM-compatible in any room of your house and it became an office: you might as well add a photocopier, a filing cabinet and a wall planner. Then Jony Ive and Steve Jobs produced the iMac, a computer you wanted to lick, an adornment to any room. It even had a handle; you weren't really meant to carry it, but it made it feel like something you could comfortably touch. Apple matched acceptable performance while changing the question to how it feels to use. Such emotional factors rarely appear in research, yet they are strong drivers or brakes.

“If you put a grey IBM PC in any room in your house, you might as well put a photocopier and a filing cabinet in the room.”

The point: When an industry competes on measurable specs, shifting the dimension of competition to feel and context can beat incumbents.

Variants: Rory frames Jobs as a reverse benchmarker, and one telling adds the iPhone's little bounce at the end of a scroll: trivial per use, enormous across a billion users a hundred times a day.

Rory often says he works in advertising because it is one of the very few ways to monetise ADHD. In logistics, sitting in a Paris café or watching a beagle on a train doesn't make you better at your job; in marketing it does. He has no sense of proportion and fixates on the most trivial element of any discussion, which means 90 per cent of the time he is a pain in the arse, but sometimes he is unusually right. He likes that advertising has let him stay himself rather than bury his quirks, and worries that by medicalising ADHD we may be treating creativity as a disorder, though an air traffic controller might be better faintly medicated.

“I work in advertising because it's a great way to monetise ADHD.”

The point: Some fields reward the wandering, disproportionate attention that others penalise, so cognitive diversity is an asset to be placed well rather than suppressed.

Variants: One telling adds an academic's finding that groups including non-neurotypical people make better collective decisions; another links it to 'Jack of all trades, master of none, is oftentimes better than master of one'.

A colleague gave Rory a perfect example of reverse benchmarking. The founder of Brazilian jiu-jitsu wanted to invent a new martial art, which is hard after a thousand years or more of refinement in the East. Then he noticed that every existing martial art teaches you to fight standing up; once you were on the ground you had basically lost, so nobody had thought about it. Fighting lying down was terra incognita. That, Rory jokes, is the more valuable skill in a British pub car park scrap while someone shouts 'Leave him, John, he's not worth it'. The founder established it by challenging other martial artists, who enjoyed a brief advantage until they were knocked over.

“No one's produced a martial art for how you fight when you're on the ground.”

The point: Unexplored space often lies in the situation everyone else treats as already lost or unquantified, not in the dimension they have all optimised.

Variants: He pairs it with Apple asking why computers had to be ugly and hard to use while rivals competed on specs, and with Uber, where everyone focused on the taxi and the magic was the map.

Rory titles talks 'Innervation', and insists it is not a typo: invention that relies on understanding the brain rather than the laws of physics, caring less about the properties of silicon and more about the properties of neurons. There are two ways to create economic value: find out what people want and a clever way to supply it, or work out what you can supply and a clever way to make people want it. Economically they are equally valuable, yet we spend 90 per cent of our effort on the first. A technical innovation can be copied in months, but a psychological one endures, because we act on feelings, not data. So the next revolution may not be technological at all; it could be psychological.

“The next revolution may not be technological at all; it could be psychological.”

The point: Changing how people perceive and value something creates as much wealth as changing the thing itself, costs less, lasts longer and is far less explored.

Variants: One telling grounds it in the Austrian School's refusal to separate objective from subjective value; another notes agencies were paid by media commission, so problems needing no bought media never reached them.

Alphabet's X, run by Astro Teller, hunts moonshots: making something ten times faster, cheaper or safer, because 3 or 5 per cent improvements rarely change behaviour. Rory agrees with the 10x aim, but Google are engineers, not psychologists, and in the physical world tenfold gains keep hitting the laws of physics and diminishing returns. Making a plane ten times faster than a 787, or a train ten times faster, is very hard. Making a train journey ten times more enjoyable is weirdly easier, and quicker, cheaper and lower-risk, because you tinker with perception rather than reality. Because psychology is messy, magic is possible. So most R&D for 10x projects should hunt for things that feel ten times different rather than are ten times different.

“Maybe you can't make something ten times faster, but you can make it ten times more enjoyable.”

The point: When physical improvement hits diminishing returns, order-of-magnitude gains are still available in subjective experience, and they are cheaper to find.

Variants: One telling notes that a ten-times cooler hotel costs a fraction of a ten-times faster plane, and that a ten-times more enjoyable train journey may be worth more than one 90 per cent shorter.

Asked whether behavioural economics is just anecdotes about quirks, Rory defends the word. Most really significant information emerges first in anecdotal form, not in data: someone notices something strange and decides to find out what it means. Nearly all discoveries, penicillin, Viagra, gravity, Darwin, start that way, and serial killers get caught because of an anecdote. Tell an entrepreneur a weird thing happened and they hear a cash register; tell a bureaucrat and he doesn't want to know, because it messes with his model of the world. Insights come from the edges, not the aggregates, and big data is a rear-view mirror. He would rehabilitate 'opportunistic' too: small businesses work intuitively, while large ones must pretend to be rational to justify themselves.

“Insights come from the edges, not from the aggregates.”

The point: Anomalies and anecdotes are the early signal of truths that aggregated data has not yet caught, so dismissing them means missing discoveries.

Variants: One telling cites Jeff Bezos: when the anecdote disagrees with the data, it is usually the anecdote that is true.

One of Rory's formative influences was a childhood book of Aesop's fables. Aesop, probably a slave living before Plato, was in a sense the first behavioural scientist: sour grapes and the dog in the manger are psychological quirks. Jesus was a pretty good one too. The parable of the lost sheep is basically loss aversion, and the parable of the vineyard is about perceived equity, arguably proportionality bias. There is a book arguing Jane Austen was a game theorist; his third book, he jokes, will be 'Jesus Was a Behavioural Economist'. The tragedy is that this knowledge was lost. Described by Aesop, Plato and Jesus, and explained by Darwin, it could found a new science with more promise than much heavily funded technology.

“The parable of the lost sheep is basically loss aversion.”

The point: Ancient stories encode behavioural insights that science only later named, so old wisdom is an underexploited source of ideas.

Variants: One telling adds the parable of the sower as systems thinking.

At a Buckingham Palace banquet in the early 1950s, with Churchill back as prime minister and the young Queen on the throne, a footman reported that a distinguished guest had pocketed a solid gold George III salt cellar. Two bad options: accept the loss, or accuse her and cause a diplomatic incident. Churchill said leave it to him. He slipped the matching pepper pot into his own breast pocket, waited until after dinner, sidled up to the guest by a window, showed it to her and said: I think we've both been spotted, we'd better put these back. Both items were returned. By posing as a fellow offender rather than an accuser, he turned a confrontation into shared mischief, and nobody lost face.

“I think we've both been spotted. We probably better put these back.”

The point: Many apparent trade-offs are emotional problems that a psychological reframe can dissolve, letting everyone save face.

Variants: The year shifts between 1952 and 1954, and the stolen item between salt cellar and pepper pot. One telling ends with Kevin Dutton asking whether a solution like that exists for practically everything.

Rory suggests anyone redoing their kitchen should install two dishwashers, and audiences look at him as if he's in the pay of Bosch. But with two you never unload a dishwasher and lose no storage. The clean one serves as a cupboard: take plates from it, eat, and put them in the dirty one. When the dirty one is full, run it, and it becomes the clean cupboard while the other, now empty, takes the dirty things. You just move things from one side to the other; you can even buy 'clean, dirty, running' signs on Amazon. Anyone without two, or without an evangelist friend who has them, thinks you're barking mad. It is obvious in retrospect and completely non-obvious in advance.

“It is absolutely obvious in retrospect but it's completely non-obvious in advance.”

The point: Many good ideas look absurd beforehand and obvious afterwards, so they are only found by people willing to question an object's assumed purpose and try them.

Variants: Asked whether he has two himself, he admits his wife refused, citing counter space.

School teaches that there is a single right answer, that the question contains all the data you need to reach it, and that the question is the question. Real life is not like school. To become a creative problem-solver, Rory says, you must unlearn three things: there is more than one right answer to any question; the opposite of a good idea may be an even better idea; and it is not cheating to change the question. Good mathematicians know this. A problem may persist not because it is hard but because everyone has asked the wrong question about it. Most entrepreneurs succeed not by answering a question better than competitors but by asking a different one, and almost every disruptive business boils down to that.

“Unlike high school maths, it's not cheating to change the question.”

The point: Reframing the question is legitimate and often the best way to solve a problem that has resisted direct attack.

Variants: One telling adds that it is also legitimate to answer a question nobody has asked yet, and points to John Kay's Obliquity.

In Mathematica, the French mathematician David Bessis studies how the best mathematicians actually solve problems, and none of them work logic-first. They start with a conjecture, almost always reached intuitively, then stress-test it with mathematics. They say they use logic not to arrive at answers but to correct their dreaming when it goes wrong. Rory concludes that rational calculus should be deployed as a sense check at the end of the process, not allowed to define the process itself. That is exactly what finance should be doing: stress-testing people's intuition, not determining what they think about. Yet business prizes quality of argument over quality of outcome, so people would rather reach a justifiable decision than one that works.

“Logic actually isn't there to help you think. It's to correct your dreaming when it goes wrong.”

The point: Generate ideas by intuition and use logic to stress-test them, rather than letting analysis decide what gets thought about.

Rory's 'creative opportunity cost': there are far more good ideas we can post-rationalise than good ideas we can pre-rationalise. Demanding that someone wins the argument before trying anything is too big an ask, and it throws away most good ideas, which only make sense afterwards. After 25 years in a creative department he agrees that ideas float in the air and you catch them, develop a nose for what is interesting, then explain it. Biology is a science of exceptions: Darwin noticed finches with different beaks on nearby islands and asked why. Data also skews to the past and to what happens fast. So test stupid things, and when you test something, test its opposite too.

“There are far more good ideas we can post-rationalise than good ideas we can pre-rationalise.”

The point: Requiring advance justification filters out most good ideas, so experiment first and explain afterwards.

Variants: One telling adds a Silicon Valley investor who looks for an element that is slightly stupid, Scott Galloway's 'stupid until they weren't', and Tony's Chocolonely succeeding partly because its founders did not know what they were doing.

Turkish barbers flick burning methylated spirit at your ears to singe off the hair. No customer would ever have asked for it. Nobody in market research ten years ago would have said: I quite like my barber, I just wish he'd flick burning methylated spirits into my ears. There was literally no demand until it happened. Yet it is the flourish you notice, remember and find special, perversely enjoyable precisely because it is gratuitous. Be totally customer-led and you lose the capacity to surprise. Such touches are easy if you have the freedom to be silly, as in a founder-led or privately owned business; in a PLC where every cost goes through procurement they are incredibly hard.

“I really like my barber, I just wish they'd flick burning methylated spirits in my ears.”

The point: Research cannot discover delighters that nobody could imagine requesting, so organisations need the freedom to try gratuitous, silly touches.

Reading Sherlock Holmes obsessively shaped Rory. Holmes says the grand thing is to reason backwards, yet people rarely practise it: fifty can reason forwards for one who can reason backwards. Forward reasoning, this is true so we should do that, is prized and taught because it involves no subjectivity, so you cannot be blamed. Reasoning backwards asks: this happened, so what conditions could have led to it? Or, as Roger Martin puts it, what would have to be true for this to come about? That is abductive inference, and it needs imagination. Holmes is presented as supremely rational but is spectacularly creative: noticing the dog that didn't bark in the night-time is a leap of imagination, not deduction.

“What would have to be true if we want more people to do this?”

The point: Working backwards from an observed or desired outcome to the conditions that would produce it opens options that forward logic never reaches.

Variants: One telling widens it into Rory's half-serious theory that cops are better scientists than scientists, and criticises political movements for fantasising ideal societies instead of asking what would have to be true.

Few people keep their student essays, and those that survive in garages are of no value to anyone. The value was never in the essay but in the process: the questions asked, the books read, the ordering of thoughts into a thesis. AI essays may be better and faster, but they are worthless, because they shortcut from request to finished product and bypass the part that mattered. Rory fears the same in advertising: weeks spent asking who we are talking to, why they should care and what makes us distinctive may be worth more than the resulting ad. Like a cruise ship, nobody boasts of its speed, because the time on board is the point. Some things gain their value from slowness and effort.

“The value of the essay wasn't in the essay, it was in the process you had to go through to write it.”

The point: Automating straight to the output can destroy value when the effortful process, not the product, was what mattered.

Variants: He also applies it to his own fortnightly Spectator column, and often follows it with Jeremy Bullmore's advice to produce an ad campaign you never run.

Viagra was being tested as a treatment for angina. Partway through the trial it clearly wasn't working, so the trial was abandoned and participants, among them a group of Welsh miners, were asked to send back their unused pills. Unlike every previous trial, many refused. Someone was clever enough to think: that has never happened before, let's find out what's going on. A useless angina cure turned out to be the right answer to a different question, a cure for which there was no known disease. Rory says a German company reportedly got there first, noticed the side effect and rejected the drug as a failure. It is perfectly acceptable to set out to solve one problem and accidentally solve another.

“It's the right answer to a different question.”

The point: Unexpected side effects and anomalies can be breakthroughs if you reframe the question instead of rejecting the result.

Variants: Another version has nurses noticing trial participants sitting in a really weird way.

Peter Sutcliffe, the Yorkshire Ripper, was caught partly because of how he parked. In Bradford's red-light district the etiquette was to park facing the wall, so nobody could see into the back seat. Sutcliffe had parked facing outwards, as if planning a quick getaway. Two beat officers noticed, because police are trained to reverse into bays themselves; Surrey Police told Rory they even do it at Sainsbury's. You cannot imprison someone for life for parking in a funny way, but it was significant enough to warrant further investigation. Rory uses it to show why police separate an investigative phase, where trivial anomalies tell you where to look, from an evidential phase, where you prove the case.

“You can't imprison someone for life for parking in a funny way, but it was significant enough to warrant further investigation.”

The point: Trained attention to trivial anomalies directs inquiry towards what systematic evidence-gathering would miss.

Variants: Often paired with Son of Sam, caught through a parking ticket, as proof that cops are more scientific than scientists.

Much of what futurologists say is nonsense: they spot a Copenhagen shop making furniture from recycled rhino dung, call it a massive trend, and six months later it's bust. But one told Rory something brilliant: there aren't really trends, there are vectors. We say X is moving to Y, but more often X moves to Y and to the opposite of Y at the same time. Media has gone both very long, three-hour podcasts and box-set binges, and very short, TikTok, while conventional entertainment sat in the middle. A very rational strategy often puts you right in the middle with all your competitors, when the direction of travel is centrifugal. Dyson assumed people would pay £600 for a vacuum cleaner, and they did.

“There aren't trends, there are vectors.”

The point: Markets often polarise, so look for the counter-movement alongside any trend and beware the logical middle position, which may be what is dying.

Variants: One telling adds that podcasts' success exposed broadcasters' failure to value conversation, which they had boxed into half-hour and hour slots.

Roger Martin, one of Rory's business gurus, points out that there is no data about the future. In finance, compliance and much of business, people say they cannot act without data to prove it. That sets the bar ridiculously high, because all big data comes from the same place, the past, and anything genuinely new has no supporting data at first. If you can only license what you do in the future on the basis of data from the past, you are creatively limited: you cannot really grow or differentiate. Efficiency is rarely questioned because it sounds logical, but it optimises for the past. So your approach must be partly experimental. Innovators, like police, are not just assembling evidence; they are creating it.

“All big data comes from the same place, the past.”

The point: Decisions about the future need experiment and creative reasoning, because data can only describe what has already happened.

Variants: One telling adds Martin's three sources of creative strategy, anomalies, analogies (Alien pitched as Jaws in space) and showing an assumed trade-off is not one, with humour as Rory's fourth. Mark Read adds that the better you model the past, the worse you may understand the future.

TRIZ was a Soviet-era innovation method that reverse-engineered Western patents to find recurring patterns of invention, treating most problems as contradictions; its originator was rewarded with years in the gulag. One pattern is to make the static thing move and the moving thing stay still, as online retail reverses who travels, shopper or goods. Rory realised that what he practises is a psychological TRIZ: sometimes a contradiction can be resolved in the mind. People hate waiting for taxis and you only have ten taxis, but let them watch the taxi approach and the wait becomes far less painful. He believes a collective effort could catalogue perhaps 27 recurring patterns shared by 90 per cent of truly innovative marketing ideas.

“Sometimes you can resolve a contradiction psychologically.”

The point: When a physical trade-off cannot be broken, changing perception can dissolve it, and the recurring patterns of such moves could be catalogued and taught.

Scientific funding usually demands a clearly identified problem and a single destination before you may experiment. Yet Andre Geim discovered graphene with equipment from a stationery shop, endlessly pulling apart pencil graphite with sticky tape; he also levitated a frog. He says grant-application science bores him and he prefers hit-and-run, drive-by research: find an interesting angle, mess around for a bit, see where it gets you, and discover the use later. Rory argues for random, curious and mischievous science that works backwards, a punt first and a rationale afterwards, and worries about a world where you must win an argument and get approval before you can act.

“You just find an interesting angle of inquiry and piss around for a bit and see where it gets you.”

The point: Exploratory, unjustified tinkering produces discoveries that goal-first funding would never permit.

Variants: One telling adds the mRNA vaccine scientist demoted at Penn who nearly quit, and David Ogilvy's admiration for Niels Bohr because his scientists played practical jokes on each other.

Rory says Australian and New Zealand economists, such as Nicholas Gruen, whom Martin Wolf called the greatest economist you've never heard of, and Steve Keen, are genuinely better. Like the cassowary, the kangaroo, the koala and the fictional drop bear, they evolved at a distance from everything else, free of the fashions of American freshwater and saltwater economics and not prisoners of standard Chicago School doctrine. The isolation that drives Australians crazy is the place's most valuable feature: you are not endlessly infected by everybody else's ideas.

“You're not endlessly infected by the ideas of everybody else.”

The point: Intellectual isolation from the mainstream protects thinkers from orthodoxy and breeds original ideas, as geographic isolation breeds unique species.

Behavioural economics, Rory says, got started largely through a happy accident: at Stanford, sometime in the 1980s he thinks, the psychologists, Tversky and Kahneman among them, happened to share a coffee room with the economics faculty. The interplay between the two disciplines was far more interesting than either in isolation. As so often, the best ideas now emerge not within disciplines but at the intersections between them. That is why Nudgestock draws people from such diverse backgrounds, and why breaking down silos matters: a pattern that makes perfect sense in one context can be borrowed to explain something apparently completely different.

“The best ideas nowadays don't emerge within disciplines, they emerge at the intersections between them.”

The point: Breakthroughs come from collisions between disciplines, so organisations should design for cross-silo contact.

Behavioural science is not a prescriptive science, Rory says: it doesn't tell you what to do, it tells you where to look. Its principal value, and the reason it isn't always welcome, is that it expands the solution space, so there are ten places to look rather than two. Economics typically offers two places on incentives: bribe people for doing something or fine them for not doing it. If someone breaks a compliance rule, an economist says fine or fire them; he would never suggest making them wear a hat with a penis on it at work for a few days, which might work better. That narrowness is bizarrely popular in business and politics because it protects decision-makers from blame.

“It doesn't tell you what to do, it tells you where to look.”

The point: The value of psychology lies in widening the range of possible solutions beyond the bribe-or-fine options that economic thinking offers.

Rory has stayed at Ogilvy since 1988 partly from curiosity: an ad agency, like consumer capitalism generally, is the Galapagos Islands for observing human behaviour, with all the anomalies and weird patterns in one place. Go out, look for peculiarities and oddities, and spot their significance. Everybody is trying to be like Newton, looking for a theory of everything that makes the world look like physics; you need to try to be like Darwin, because the only correct response to a complex world is observation, often of anomalies. Practitioners are often far ahead of academics because they stumble on discoveries, but they fail to codify them. He wants a symbiosis: practitioners find the oddities, academics mathematise and test them.

“Everybody's trying to be like Newton, and you need to try to be like Darwin.”

The point: In complex systems, collecting and explaining real-world anomalies beats imposing grand theories, and markets are the richest place to find them.

John Cleese's book on creativity cites 1950s Berkeley research comparing ten first-rate architects with ten workaday ones, none told why they had been chosen. Their habits varied, night owls and early birds, but one thing split the groups perfectly. The indifferent architects all started drawing straight away; the creative ones did not put pen to paper immediately. They spent time doodling, playing, waiting for inspiration and perhaps re-asking the question. Rory reads this as being emotionally content in a space of not knowing, as Margaret Heffernan describes, and waiting to get lucky. His wife wants to know what is happening on Saturday; he would rather wait and see. As John Lennon said, time spent doing nothing is rarely wasted.

“You are content in a space of not knowing.”

The point: Tolerating uncertainty before committing to a solution leaves room for better ideas to arrive.

One reason people find creative types annoying is that they have no sense of proportion: they obsess over trivial details and ignore things they are supposed to care about. Rory argues they are right, because their audience has no sense of proportion either. Real customers respond disproportionately to small things and are indifferent to big ones. One phrase said two slightly different ways, trivial to a grammarian, can arouse diametrically opposite emotions. Drawing on David Snowden's Cynefin model, he says procurement, logistics and operations people build careers in the simple and complicated domains, while marketers work in the complex one, which is why the former find marketers maddening.

“Creative people are right not to have a sense of proportion, because neither does their audience.”

The point: In complex human systems small differences are not proportionately small in effect, so an eye for the trivial is an asset.

Rory watches a clip of Richard Feynman roughly every two weeks for fun. Feynman is explaining how a new scientific law is found: 'First, you guess.' The lecture hall bursts out laughing, and he says don't laugh, that is exactly how it works. Everything starts with a guess, something from the unconscious. Then you compare it with experiment, and if it disagrees with reality, the guess is wrong. Everything begins with some degree of speculation; logic and experiment are the check, not the source.

“Don't laugh, that's exactly how it works.”

The point: Discovery starts with intuition and uses logic and experiment to test it, so demanding logic first gets the order wrong.

Variants: One telling adds Rory's aside that Feynman's last wife was British, probably because only British women have low enough expectations of male behaviour to put up with him.

On the day of the discovery, Alexander Fleming was not supposed to be at St Mary's. He was meant to be away at his cottage, but a crisis brought him grudgingly into London, and while there he looked at his Petri dishes. Had he decided that one simply hadn't grown properly, he would have thrown it in the bin with the rest. Biology is a science of exceptions, not laws, and the anomaly most people try to freeze out because it is annoying is really where the news comes from. Once you succeed, nobody cares how you got there: Rory asks CEOs whether anyone refuses penicillin because its discovery was not planned in advance.

“If there's anybody here who refuses to take penicillin because its discovery was not planned in advance, you're very principled, but quite likely to die.”

The point: Breakthroughs hide in the anomalies that tidy-minded people discard, and an accidental discovery is worth exactly as much as a planned one.

Variants: One telling turns it into a joke about return-to-office mandates, then notes Newton discovered gravity working from home during a pandemic: swings and roundabouts.

A UN agency was given a huge donation of children's shoes by Crocs, for parts of Africa where children lacked shoes. The logical plan, which 99 per cent of people would follow and nobody would criticise, was to find where people had no shoes and hand them out. Instead the person in charge asked how to get double duty from them: give a pair to every child who turned up at school in the first week. Children who saw others with new shoes asked where they got them, and turned up the next day so as not to miss out. It shod the unshod and created the habit of attending at the start of term, which makes continuing far more likely. That is alchemy: a free lunch.

“If you use Newtonian physics there's no free lunch. If you look at the world psychologically there are free lunches everywhere.”

The point: The same resource, delivered with a little creative conditioning, can achieve a second goal at no extra cost.

Variants: The agency is UNESCO in one telling and UNICEF in the other. One adds Rory's unhelpful joke that Crocs are a fantastic contraceptive; the other his line that it is much easier to get fired for being illogical than for being unimaginative.

A client wanted commercial innovation, the kind that changes the relationship with the consumer rather than the product, like Amazon Prime or Share a Coke. They proposed to define it, build a framework and run a process to select the best ideas: theorise first, have ideas second. Rory asked why not do it backwards: have a load of ideas, then post-rationalise them, because that is how it really works; we just pretend it isn't. Pre-rationalisation mostly feeds senior egos that want to believe good things come from the top, whereas most progress is bottom-up. Nobody could get married or choose where to live that way. You couldn't do anything interesting in life if you first needed a 56-page PowerPoint deck.

“Why don't you just have a load of ideas and then post-rationalise them, because that's how it really works.”

The point: Defensive, theory-first process substitutes for experimentation and filters out interesting ideas; generate first, justify after.

Jeremy Bullmore always said that even a company with no media budget and no intention of advertising should produce an advertising campaign for itself, and Rory thinks very many medium-sized B2B firms should. The value lies in the questions it forces: why are we here, why do we deserve to exist, who do we serve, what makes us different, what do people pay for that we provide too little of? Those questions and the journey to the end point deliver perhaps 80 per cent of the value, even if the ads never run. Leap straight to the finished ad and you lose that. In business the by-product can be more valuable than the product, so be careful what you shortcut.

“The process of producing an ad for a company forces you to ask questions like what are we for, who do we serve.”

The point: When the value lies in the questions a process forces, shortcutting to the output destroys it.

Variants: Usually paired with the worthless student essay: the essay had no value, the work of writing it did.

Drug companies rationally make pills as small and easy to swallow as possible. Rory and Dan Ariely, hearing this on a call, worried: something tiny and effortless is also forgettable, and the biggest problem in treating many conditions is patient compliance. For a drug taken every day it may be better to make it awkward: grind it with a pestle and mortar and dissolve it in water. A ritual is hard to forget; after a week you notice when you haven't done it, whereas with three tiny pills you can't remember whether you've taken them. More effort in preparation would probably boost the placebo effect too. Advertising, he says, is the Galapagos of behaviour: oddities found in trivial settings apply to big problems.

“When you make something very small and very easy to take, you also make it very forgettable.”

The point: Deliberate friction can create a memorable ritual that improves adherence and perceived efficacy, against the instinct to make everything effortless.

Rory reckons perhaps 90 per cent of progress, in business, science and pharmaceuticals, is messy, non-linear and happens backwards. Penicillin and Viagra came from someone making a lucky observation, asking what it meant and building on it. Darwin didn't start with an idea; he looked at birds and worked it out backwards. Scientific progress is much more penicillin than sequential logical steps agreed in advance. Yet in an organisation that must justify everything by cost-benefit analysis, you must do the reasoning first, and most effort goes into pretending upwards that things make more sense than they do. That pretence has an opportunity cost: Red Bull or miso soup make no sense until recontextualised, and bottom-up ideas never surface.

“Scientific progress is much more penicillin than it is sequential logical activities all agreed in advance.”

The point: Discovery usually precedes explanation, so demanding a business case first kills most breakthroughs.

Nobody went to James Dyson saying they were annoyed because they were looking for a chance to spend more money on a vacuum cleaner, and no data point would have told you they were. Had Dyson pitched an £800 vacuum cleaner to Rory in the 1990s, he says he'd have explained that the market peaks at about £400 for a Miele; that it's a distressed purchase you make when you leave home or the old one breaks; that no couple goes vacuum cleaner shopping for the lols; and that anyone who can afford £800 pays someone else to clean. All perfectly good logical objections. Had Dyson returned with a £400 hairdryer, Rory would have had him escorted out as a lunatic.

“There was no data point that was going to tell you that.”

The point: Breakthrough propositions are invisible in data and fail logical scrutiny in advance, so demand for them has to be discovered by experiment.

Variants: One telling pairs it with Starbucks: nobody complained that they couldn't spend $5 on a coffee. Another cites Roger Martin: big innovations come from 'what if', not 'what is'.

In most businesses, accounting or law, rationality is the gold standard: sound rational and you win the argument. In advertising, hospitality, theoretical physics or entrepreneurship, Rory argues, rationality is the bronze standard. A rational solution is fine; you won't get fired. But present a rational ad and you hear: yeah, it's all right, can you do better, funnier, more memorable? Silver is ingenuity, and gold is complete magic, and all magic relies on psychology; even technology is only magical because of how it makes us feel. He is not against data or reason; his complaint is that accountants, lawyers and economists have a monopoly on decision-making. If you don't believe in magic, you won't look for it.

“If you don't believe in magic, you won't look for it.”

The point: Treat the rational answer as the floor and a springboard, not the ceiling, or imaginative solutions will never be sought or recognised.

Variants: One telling grounds it in John Ralston Saul's Voltaire's Bastards: reason is only one of the brain's capacities alongside imagination, creativity and common sense, yet we made it the gold standard.

Rory describes stories as the PDF files of human information: the universal format for storing and sharing information that any recipient can open, whatever their hardware. Just as you convert a document to PDF so others can read and sign it, you must get an idea into story form for it to be memorable and shareable. Nobody stays in the pub while someone recites data or shows graphs. Asked whether good stories need embellishment, he recalls David Ogilvy, challenged on the accuracy of a story in his autobiography, replying in effect: I know, but it's a good story, so shut up.

“Stories are the PDF files for the human mind.”

The point: Ideas only travel between people once converted into story form, so delivery matters as much as content.

Variants: In another telling it comes while crediting his Welsh upbringing for his speaking style, alongside Taleb's advice that slight difficulty in following a speaker makes listeners pay more attention.

Rory's confession after 30 years in advertising: agencies pretend there is a linear process for generating ideas, but there isn't. Checklists are valuable, but the process, in as far as there is one, is basically talking nonsense until you get lucky; that is about half of how it works. More than half of great campaigns started with an idea and worked backwards to a brief. The pretence exists to keep procurement happy, since agencies charge by the hour and can hardly ask for thousands of pounds to talk until they get lucky, and to preserve senior people's status. Nobody planned you either: you are the product of lucky accidents, ancestors who managed to have sex before they died.

“We can't go and say can you give us thousands of pounds to talk until we get lucky.”

The point: Creative work is partly random and bottom-up, and forcing it into the appearance of a linear process filters out good solutions.

One of the greatest brand assets imaginable is the Dulux dog. You can post-rationalise why an Old English Sheepdog suits a paint brand, but it didn't happen like that. The crew were shooting a freshly painted room in a rented house whose owner had an Old English Sheepdog. It escaped and wandered into shot, and the director, astute enough to see it looked a lot better with the dog in it, said the dog stays in the picture. That accident became literally a billion-dollar brand property. Much of advertising's success is accidents plus someone with the perceptual acuity to notice their significance and the confidence to act. If everything must be pre-scripted by rationalist accountants, the creative solution space shrinks to a tenth.

“That is literally a billion-dollar brand property.”

The point: Huge assets often arise from accidents, so the skill lies in noticing them and having the nerve to keep them rather than pre-scripting everything.

Variants: One telling credits the director, the other the photographer; Rory also uses it as proof of fat-tailed outcomes when Ogilvy brings the dog back.

One creative technique Rory uses is what he calls the intermediate impossible: impose an absurd constraint to force you out of the default solution. For example: we have to make money from an airline, but the one limitation is that the planes are not allowed to leave the ground. Riding to a studio, he applied it to London taxis: they should earn money while stationary by renting out the space. Every cab could have a small table; you call one down and take a Zoom call or work in it for an hour while the driver has a meal. Otherwise, in London, you end up buying three coffees in a café while hunting for a quiet corner.

“We've got to make money from an airline, but the one limitation is the planes aren't allowed to leave.”

The point: An impossible constraint forces thinking away from the default use of an asset and reveals hidden sources of value.

Of the three routes to a better future, technology, economics and psychology, Rory is most optimistic about psychology, for a simple reason: we are much more wrong about psychology than we are about physics, so the potential for progress is greater. Behavioural economics has arguably advanced more in the last 20 or 30 years than in the previous century. Producing a faster aircraft is going to be really hard; producing an aircraft that people enjoy being on is an easier problem. We spend disproportionate attention on objective reality when we should be experimenting with context, meaning and perception, improving the human condition within our minds rather than through external invention and consumption.

“We're really much more wrong about psychology than we are about physics.”

The point: The biggest untapped gains lie where understanding is weakest, which today is human psychology rather than technology.

IBM's endline 'Solutions for a small planet', which Ogilvy gave them in the late 1990s, came out of a research group where someone said it as a criticism: all you people are doing is coming up with solutions for a small planet. Someone in the room heard it and took it. A lot of creativity isn't origination but inspired observation. Rory likes the parallel of owning an insult: Welsh rugby fans bring inflatable sheep to matches, and Wisconsin fans wear cheese wedges on their heads.

“A lot of creativity is just an act of inspired observation, nothing more, nothing less.”

The point: Good ideas are often found by listening, including to criticism, rather than invented from scratch.

A lost salary is quantifiable and concrete; an opportunity cost is nebulous. That is why Rory says 20 per cent of your life should be deliberately a bit randomised. His own employer sometimes asked, 'Why are you going to this conference?', and he would answer, 'I don't know yet, because I haven't been.' A proportion of your resources should go to exploratory activities, because focusing purely on the day job feels efficient while hiding what you miss.

“I don't know yet because I haven't been.”

The point: Exploration cannot be justified in advance, so a fixed share of time should be ring-fenced for it.

London Bridge station got a billion-pound rebuild but forgot the one bit of 'magical bollocks' St Pancras had. Rory would have had two shops there not to make money but to be cool: Europe's largest florist, a carpet of flowers at the foot of the escalators, rent-free if it kept long hours, so everyone would say the flowers are stunning. The finance director would kill the idea and the florist first. An engineer even objected to earlier boarding because it lost station retail sales.

“Just do something really, really weird.”

The point: Spend a sliver of a big budget on something memorable rather than purely profitable.

Rory cites Paul Feyerabend, who called himself a methodological anarchist [?]: worthwhile discoveries have not all come from obeying strict scientific method, so why let methodological purity restrict your solution set? Science should be useful, not a status-signalling vehicle for obsessive rationalists. We must become comfortable with progress that arises from random accident, rather than letting those good at post-rationalisation monopolise the toolbox. Most evolutionary progress happened by fluke.

“A good guess which stands up to empirical observation is still science.”

The point: How an idea was found matters less than whether it works when tested.

When a viewer objected to a speaker's use of left brain and right brain, Rory defended it. A metaphor, to be useful, does not have to be true; it can simply be useful. And he points the sceptic to the neuroscientist Iain McGilchrist's work on the hemispheres. The shorthand for rational, institutional thinking versus intuitive, creative thinking earns its keep regardless of the anatomy.

“A metaphor to be useful doesn't have to be true.”

The point: Judge models and metaphors by how well they help you act, not only by their literal accuracy.

Rory had just discovered the American logician Charles Sanders Peirce, who argued that most problem solving is not sequential thought but 'abductive inference'. You notice something unusual, or imagine something different, and then work out what the prevailing conditions must be for that to be normal. That is what Darwin did: all these finches have different beaks, so what would have to be true for that to emerge? It always requires a leap of imagination, so any process that does not allow for that leap is flawed from the outset.

“Anything that doesn't allow for that leap is essentially flawed from the outset.”

The point: Real problem solving starts with an imaginative leap from an anomaly, which purely sequential logic cannot supply.

Rory argues The Wealth of Nations, like The Origin of Species, has been unfairly cherry-picked down to a single idea, and calls Smith the world's first behavioural economist. One of his favourite sentences describes people 'whose trade it is not to do anything, but to observe everything, and who upon that account are often capable of combining together the powers of the most distant and dissimilar objects'. Two hundred years before TRIZ, that is a brilliant theory of innovation: a recombinatorial process of taking something from one place and melding it with something completely different.

“Their trade is not to do anything, but to observe everything.”

The point: Innovation comes largely from observation and recombining ideas from distant fields.

Digging into Aesop, Rory found many fables that are really nudge experiments. A farmer leaves his land to his children and tells them treasure is buried there. They dig up the whole farm looking for it, and in doing so they plough it. They do find treasure, in the form of greater crop productivity. He also notes Jesus's parables, such as the lost sheep and loss aversion: people thousands of years ago wrestled with perception versus reality and illustrated it anecdotally.

The point: A motivating story can get people to do the useful thing they would never do for its stated reason.

Rory uses AI as a checklist: it generates a good list of "have you thought of this, have you thought of that". There is never anything wrong with checklists; they are wonderful in surgery and aviation and also in creative work, much like Ogilvy's use of the MINDSPACE framework, because whole avenues of inquiry are otherwise overlooked. And AI has a "temperature" setting: turn it down for plodding, low-risk output, or up to levels of weirdness that could be inspiring. Synthetic research also lets you question the same respondents again and again, which would annoy real humans.

“There is never anything wrong with checklists.”

The point: AI is useful creatively as a prompt against overlooked options and a source of controlled randomness.

Rory asked an AI to explain a crude pub-name joke he told his children years ago, and it understood it perfectly, if primly. Asked for a similar joke, it produced a nonsense pun about a Japanese restaurant that made no sense at all, as if it had slotted random words into a joke structure. If you treat a joke as emblematic of a big creative idea, AI cannot yet originate one, though it may be a good sidekick and could amplify an idea once grasped. AI can amplify brilliance or replicate mediocrity, and he would use it to shrink admin, not the people doing valuable work. Agencies should also stop charging by the hour.

“AI can explain jokes very well, but it can't come up with a joke on its own.”

The point: AI is better at exploiting existing ideas than at exploring for genuinely new ones.

Discussing short sellers, Rory notes that they look for what everybody else believes, assumes or tacitly takes for granted. The interesting question about AI is whether we will use it to reinforce our confirmation bias or for what he calls blind-spot detection: asking what we are all ignoring. The Big Short, his near-favourite film, is about people who saw collective insanity and misaligned incentives.

“Will we use it to reinforce our confirmation bias, or will we use it for what you might call blind-spot detection?”

The point: The most valuable use of analytical tools is finding what the consensus overlooks.

Asked how to improve client briefs, Rory starts with: have you defined the problem? AI may be interesting, but if you ask the wrong question it will just get you to the wrong answer faster. The real value is redefining the question in a more interesting way. He also thinks briefs prize linearity too much: advertising that does not answer the brief but makes you famous is valuable, yet saying 'we'll make you famous' sounds unclever, as does 'put a dog in your advertising', so agencies talk about funnel optimisation instead.

“If you're asking the wrong question, it will just get you to the wrong answer faster.”

The point: Reframing the problem matters more than speeding up the answer, and broad goals like fame deserve a place in briefs.

Traditional market research writes down a list of questions, so the questions reflect the researcher's preconceptions about what matters. With AI market researchers, Rory says, the inquiry can be run as a conversation that goes wherever it goes. The great advantage is that you can start an inquiry without preconceptions about where it will end.

“We can actually start an inquiry without preconceptions as to the end point.”

The point: Open-ended inquiry surfaces things fixed questionnaires cannot, because the questions embed assumptions.

Rory invokes the science fiction trope where aliens take over the world and then turn out to be killed by salt water or airborne microbes we are immune to. Because Covid was a new virus, he thought it not unlikely that something surprisingly banal and already in common use, like spraying something in the air, might turn out to have an extraordinary effect.

The point: Novel problems may yield to cheap, mundane solutions nobody thought to try.

You only reach an interesting answer if you start with an interesting question, Rory says, but modern business makes it too easy to be boringly, shallowly logical, what Nassim Taleb calls naive rationality: 'this makes sense, I won't get fired.' He wants a period before any project where people are allowed to be Pythonesque. Comedy is a huge influence on behavioural science because comedy notices things everybody wants but nobody talks about. You should not be a behavioural scientist without a sense of humour.

“You only come to an interesting answer if you start with an interesting question.”

The point: Absurdist, comedic thinking surfaces truths that logical framing hides.

We tend to count success as valid only if we planned it. Rory finds that too deterministic: in life you should always place a side bet, something you are not sure where it is going but are currently interested in, because it maximises your chances of getting lucky. You do not go to a party with a specific intention; you go to see what happens. Large businesses define everyone's job so narrowly, to look efficient, that they pay a hidden cost: no lucky accidents. Most breakthrough discoveries, like penicillin and Viagra, involved luck.

“You ought to always in life place a side bet.”

The point: Leaving room for undirected curiosity is how organisations and people get lucky.

Asked why he never lived in America, Rory says advertising would mean New York, and New York, LA, DC and San Francisco are annoyingly industry towns: finance, entertainment, government, software. A friend says Silicon Valley dinner parties are unbelievably boring because everyone talks tech, whereas a London dinner party mixes journalists and people discussing education policy or architecture. He also notes an asymmetry: the good stuff of the US eventually crosses the Atlantic east, but cathedrals and medieval towns do not travel west.

“Silicon Valley dinner parties are unbelievably boring because everybody just talks about tech.”

The point: Mixed environments generate more varied ideas than monocultures.

Roger Martin says real strategic breakthroughs happen like scientific ones: by spotting an anomaly, by an imaginative analogical leap such as asking what is the Red Bull of hotels, or by finding a trade-off and resolving it creatively, as in reassuringly expensive, which turns a weakness into a strength. Restaurants sometimes do the opposite, amplifying a negative such as refusing to serve ketchup as proof of authenticity. Rory adds a fourth route: humour, a non-logical gift we strangely undervalue at work.

“What's the Red Bull of hotels?”

The point: Breakthroughs follow recognisable non-logical routes that can be deliberately practised.

Rory supports flexible working partly because there is surprising value in turning off the computer and stacking the dishwasher. People say Archimedes had his idea in the bath, but he actually had it getting into the bath. In the transition between mental states the brain briefly goes into neutral, like changing gear in an old petrol car, and something may happen that would not otherwise. Kant walked at the same time every day; Nassim Taleb is a devotee of walking. Our brains are embodied.

“People always say he had the idea in the bath, but he actually had the idea getting into the bath.”

The point: Transitions and breaks are productive parts of creative thinking.

Rory is less interested in AI making decisions than in what he calls artificial inquisitiveness: make it point out interesting things to you. Setting it loose on a data set to find the anomalies may be more valuable than having it produce the generalities. He notes it is also a plagiarism engine and confabulates: asked about him, it said he had a CBE, studied PPE rather than Classics and worked for McKinsey, constructing something plausible rather than true.

“Don't make this thing make decisions, make it point out interesting things to you.”

The point: AI's best use may be surfacing anomalies for humans to explore rather than deciding.

One of the simplest ways to solve a problem is to ask a question no one has asked before. The reason it has not been asked may be that no one was clever enough, but more likely no one was stupid enough: many questions make you sound dumb because a pre-existing rational explanation already seems to cover them. Rory's twelfth rule is to dare to look stupid.

“A more likely explanation is that no one was stupid enough to ask it.”

The point: Fear of looking naive leaves obvious-seeming questions unasked, and that is where fresh solutions hide.

Rory follows Warren Buffett's 'always invert': about a third of the time his secret trick is to ask the question the opposite way round. Briefed to get more people on trains, marketers list benefits; he asks why people don't go by train. Some reasons are sensible, such as needing a car at the other end or carrying luggage, so leave those people alone. Others are dumb: it never occurred to them, car owners have a sunk cost, or someone believed trains do not stop in Bristol. Tackle the stupid objections and leave the good ones.

“Don't say why should people buy this, say why aren't they buying it.”

The point: Inverting the question surfaces silly, fixable barriers that a list of benefits would never address.

One advantage of age: if you are 28 and ask a dumb question, people think you are dumb; at 59, they think you are being profound. It is like a Buddhist monk asking 'why do people buy dishwashers?' A 23-year-old would be told 'to wash their dishes, you dummy.' Design thinking and business philosophy involve infantile-sounding questions: why do people enjoy this, do they really care or just say they do, do they even notice? Finding psychological explanations means going conversationally off-piste.

“If you're 28 and you ask a dumb question, people think you're dumb. If you're 59 and you ask a dumb question, people think you're being profound.”

The point: Naive questions unlock psychological insight, and status changes how they are received.

Hearing about Omaze's money-can't-buy prizes, Rory riffs that he has always thought you shouldn't need Tube train drivers, because everybody would love to drive a Tube train for a day. You could make it an auction prize and eventually do away with drivers, though he concedes there would be safety and training implications. He also jokes that operating a defibrillator would make a good prize, since everyone who has watched hospital dramas has a small fantasy about the paddles.

“Everybody would love to drive a Tube train for a day, wouldn't they?”

The point: Work some people are paid to do is an experience others would pay for.

Asked how he finds ideas, Rory admits his reading took a hit after an eye operation and he now uses audiobooks, but he browses online very extensively. His banal-sounding tip is to get good at using Google: brackets, plus and minus signs, inverted commas. Nobody is taught these techniques. Everyone talks about the largely imaginary class of AI 'prompt engineers', but before that we need people taught to search well, because it makes a big difference to finding things that are relevant and surprising.

“Before that we need people just to be taught how to use Google really, really well.”

The point: Basic search skill is an undervalued source of relevant and surprising ideas.

The most valuable thing about Rory's time championing behavioural science was not what he or his team did, but that it created space for the conversation. Once they had talked to a business about it, it became permissible to say: we're assuming that if we drop the price demand will go up, but shouldn't we also test putting the price up? Sometimes demand rises when price rises. To an economist price is a number; to a customer it is a feeling.

“To an economist, price is a number. To a customer, it's a feeling.”

The point: A new discipline's greatest value can be legitimising counterintuitive questions inside an organisation.

Rory says he is a huge fan of Byron Sharp and Mark Ritson, who reach many of the same conclusions from different approaches, and they are probably right where they disagree with him. He fundamentally accepts that some of the things he says are wrong. His point is that being wrong occasionally is the price you pay for being interesting, a trade-off we just have to accept.

“Being wrong occasionally is the price you pay for being interesting.”

The point: Original thinking requires tolerating some error.

Give a team two candles and a box of matches and ask them to boil some water. Shrewd people calculate that the candles lack the calorific value, tick 'cannot be done' and move on. But the boiling point depends on altitude: take it somewhere very high, or create a vacuum, and the same heat may suffice. Or move the water to France, where everybody knows water boils at 50 degrees, which is the only thing that explains the quality of the tea. Many problems are solved by recontextualising them and adding a variable nobody considered.

“You could just move the water to France where everybody knows water boils at 50 degrees.”

The point: Problems declared impossible often become solvable once you change the context rather than the inputs.

Brian Eno, whom Rory sees as an heir to Smith on innovation, argues our thinking tends to optimise along a single axis; add a second, perpendicular axis and you find unexplored space. Rory's example: billions spent increasing electric car range, when you could reduce anxiety instead. (In Britain, drive 300 miles in a straight line and you end up in the sea.) Everybody focused on what a computer could do; Steve Jobs asked what it feels like while you're doing it. Before the iMac, a computer turned any room into an office.

“Steve Jobs asked the question, 'Yes, but what does it feel like while you're doing it?'”

The point: Breakthroughs come from optimising a dimension competitors are not measuring.

Buckminster Fuller called himself a trim tab, after the small part of a boat or plane where the smallest movement has the biggest effect on the craft's direction. Rory offers it as a model for all marketing thinking: the opposite of proportionality. If you obsess about numbers you adopt a mechanistic view where things happen in proportion; marketing, at the price of some unpredictability, offers trim-tab moments where one tiny intervention has massive effects, for good or ill.

“The smallest movement has the biggest effect on the direction of the craft.”

The point: Look for the small, leveraged intervention rather than assuming big change needs big inputs.

Curiosity is decisive. David Ogilvy described the good copywriter as typically an extensive browser in all kinds of fields. Rory's tip for anyone running a creative department: buy everyone YouTube Premium, and the Audible deal of 24 audiobooks a year for about £120. Make sure they have all the mechanisms needed to feed their curiosity. Creativity, he argues, is not that scarce; it suffers from under-appreciation and under-adoption.

“If you run a creative department, just buy them all YouTube Premium.”

The point: Cheaply feeding curiosity is one of the best investments in creative output.

Rory thinks landscape gardening is one of England's least heralded artistic achievements of the last 500 years. Without people like Capability Brown, some French person would have turned the place into a symmetrical chessboard like the garden in Last Year at Marienbad. The idea that a garden should replicate an ideal form of natural landscape was a genuine creative leap of the 1776 era. (His parents-in-law garden; he advised them to take up crack cocaine instead, as it's cheaper and less addictive.)

The point: Reframing the purpose of something, here a garden, can create an entire art form.

Comedy is brilliant because it has fast feedback: there are few bad comedians, not because they do not exist, but because they quickly realise nobody is laughing. Like performance marketing, comedians tinker with the word order of a sentence to see what gets the bigger laugh, and a joke they think brilliant sometimes fails. Nobody can say why flipping the word order stops something being funny. Free-market capitalism works by the same endless trial and error, because nobody can describe in advance what they want or why they like what they like.

“Nobody can describe in advance what it is they really want or indeed why they like the things they like.”

The point: Where preferences cannot be articulated, rapid testing and variation beat theory.

Ogilvy describes its work as behavioural science and creativity together, complementary goods like a hot dog and a hot dog roll. AI can be trained on behavioural science, and Rory likes checklists, but AI tells you what to do, not how. All its training data comes from the past, so it can make incremental Bayesian progress but struggles to leap. It could understand the pratfall effect; could it have come up with 'reassuringly expensive' had that line not existed in 1974? He's genuinely not sure.

“All AI training data comes from the same place, the past.”

The point: Knowing a principle is not the same as finding the original execution that brings it to life.

Nudgestock 2021 was billed as 200,000 years of BS, and BS also abbreviates bullshit. The standard reaction is: we can't go there. The creative reaction is to run towards the fire. Given a contradiction, the typical person runs away from it and pretends it is not there; the creative person embraces it. Guinness takes ages to pour: good things come to those who wait. Stella is reassuringly expensive. Avis is not as big as Hertz: we're number two, so we try harder. It is why creatives are a pain in the arse, and necessary.

“The creative person's reaction is to run towards the fire, not away from the fire.”

The point: Owning a weakness or contradiction openly is often the most powerful persuasive move.

Rory notices a strange asymmetry in business. Creative people always have to present their ideas for assessment by highly rational, left-brain people. That is probably fine: you don't want the people who check the wheel nuts on planes to be wildly experimental. But it never happens the other way round. A room of accountants will reach a conclusion by their own formula and never suggest throwing the problem out to some wacky people to see if there are alternative solutions. He links this to Iain McGilchrist's point that the left brain has assumed a dominance it doesn't deserve.

“They'll never suggest just throwing it out to some wacky people to see if there are alternative solutions.”

The point: Organisations let logic veto creativity but never let creativity challenge logic, which narrows the solutions they ever consider.

Rory fights the idea that creativity in business should be confined to the communications function. You can be creative across all five Ps of marketing. You can be really creative with pricing: Amazon Prime is a genius billion-dollar idea. If you can solve a marketing problem with an app or a piece of choice architecture without giving millions to media owners, that is exactly what you should do. He welcomes the broader canvas, even if TV advertising's best has taken a hit.

“You can be really creative with pricing. Amazon Prime is a genius billion-dollar idea.”

The point: Creative solutions can come from pricing, product or process, not just advertising.

Outside a few R&D departments and think tanks, the only people in business or government paid to generate ideas work in creative departments, and they are usually left to the end of the process. Part of creativity's value is solving known problems, but the greater part is spotting opportunities. If creatives may only work to a brief, as advertising agencies do, that opportunity-spotting role is completely missed, because known, visible problems are a small subset of the available opportunities.

“Solving known and visible problems is a very small subset of available opportunities.”

The point: Bringing creative people in only at the end confines them to briefs and forfeits most of their value.

Asked what he looks for in people, Rory names a single quality: curiosity. The one deal breaker for employing someone in a creative role would be someone who is fundamentally uncurious. Skills can be learned, but a lack of curiosity cannot easily be fixed.

“The one thing that would be a deal breaker would be someone who is fundamentally uncurious.”

The point: Curiosity is the trait that underpins creative usefulness, so screen for it above credentials.

Over dinner, Rory and Dan Ariely debated for ages whether Ariely, planning six months in London, should live in the city or the country. Then they looked at each other and laughed: the optimum was to do both. There is no point having a big house in London, because what is the point of being in London if you are indoors? Get a crash pad in town and a place with an orchard in Kent. We automatically frame decisions as either/or, and the pretence of rationality implies there is only one right answer.

“What's the point of being in London if you're indoors?”

The point: Even behavioural scientists default to either/or; the best answer is often a combination.

When Rory's team founded Ogilvy's behavioural science practice, one mantra was dare to be trivial. It is uncomfortable for people with a Harvard Business School mindset to accept that success or failure is not always down to big things like supply chain management, but can turn on a trivial psychological effect where you hit pay dirt. The consumer brain has no sense of proportion; it uses arbitrary heuristics. One of the most valuable things in business is being gratuitously noticeable, which to the financial mindset seems absurdly trivial.

“One of the most valuable things you can have in business is just being gratuitously noticeable.”

The point: Small, seemingly trivial psychological details can decide commercial outcomes.

Much advertising works backwards: a weird idea arrives through inspiration and you reverse-engineer why it might work. Many see that as cheating, just as the German drug company saw Viagra as a failed angina pill. Rory's rule: don't apologise for a post-rationalisation. If it's an answer, it's an answer. This isn't a high-school maths exam where you have to show your working. Some Cambridge companies now do pharmaceutical research backwards, finding interesting compounds and seeing what they treat: post-rational rather than pre-rationalised science.

“This isn't like a high school maths exam where you have to show you're working out.”

The point: How an idea was reached matters less than whether it works.

We love comparison because it lies at the heart of self-justification. Rory would not try to win the Olympics: how likely is it, at 15, that you will become the world's best middle-distance runner? We see the winners, not the thousands who failed. Better to invent a new sport, like padel, at which you stand a reasonable chance. Playing by other people's rules is wonderful for spectators and a disaster for players; the hundredth-best tennis player lives out of their car. The exception is if you love the sport for its own sake.

“What you need to do is invent a new sport, like padel.”

The point: Competing on established terms favours a tiny few; creating a new arena improves your odds.

Donella Meadows, the complexity and environmental thinker, ranked the ways to intervene in a system. At the bottom: twiddling numbers, parameters, buffers and delays, the measurable small stuff that gets all the attention. Most powerful is the mindset or paradigm from which the system emerges, and above all the power to transcend paradigms. Rory thinks Ogilvy did it to a degree with Dove, and that it may be less difficult than she believed. The greatest soft power is a paradigm shift: rather than change the world, change how people see it, and their behaviour changes the world.

“Rather than trying to change the world, we can change how people see the world.”

The point: Changing the frame through which people see a problem has more leverage than tuning its parameters.

Rory recalls Edison's habit: he found the state between waking and sleeping especially rich creatively, so he would doze off holding a fistful of ball bearings over a metal plate. As soon as he fell asleep they would drop and clang, waking him; he'd pick them up and return to the same half-asleep state. Rory notes this is the opposite of the clarity and articulateness people assume is the essence of thought.

The point: Some of the best thinking happens in unfocused, semi-conscious states that productivity culture ignores.

Rory supports electric cars for a different reason from most: optionality. Whether an electric car beats a petrol one in 2025 matters to a buyer, but for policy, the internal combustion engine has limited innovation left, while electrification has unleashed a Cambrian explosion: electric scooters, cargo bikes, Citroën's tiny house-on-wheels car, the Heathrow pod, drones. Electric cars are essentially software and very reliable, so the urge to buy something eccentric like an Alfa Romeo, once suppressed by fear of breakdowns, is liberated. He bought a Lotus Eletre as a work of art.

“All the interesting things happening in ground-level transportation are a product of electrification.”

The point: Back technologies that open many future possibilities, not just those that win today's comparison.

One simple item of entrepreneurial bravery, Rory says, is actually arrogance. He tries not to be arrogant most of the time, but there are moments when being able to believe you are right and everybody else is talking rubbish is a necessary step. It doesn't play well, because most people take their opinions from each other. Another form of bravery is the willingness to make yourself look ridiculous in serious or scientific circles.

“Being able to believe that you are right and everybody else is full of it is a necessary step to take.”

The point: Breaking from consensus requires a confidence that looks like arrogance.

Point out an anomaly to someone in a large corporation and it annoys them; point it out to an entrepreneur and they get excited, seeing arbitrage. Corporate people want to know what everyone knows, to demonstrate rightness; entrepreneurs want to know what nobody knows. Rory recalls a dramatisation about the Getty kidnapping in which the family first got rich by noticing a train struggling across a patch of desert: the struggle suggested a dome in the land, which might mean oil. Entrepreneurs live in a world of discovery, not proof.

“If you're an entrepreneur, you want to know what nobody else knows.”

The point: Trivial anomalies are the raw material of discovery for those not seeking proof.

Explaining why he still likes Twitter despite Elon Musk being divisive, Rory says he indulges him on the grounds that practically everybody really remarkable in human history has been weird, difficult, a pain in the arse and annoying, and that applies to the good people as well as the bad.

“Practically everybody really remarkable in human history is bloody weird, difficult, pain in the ass, annoying.”

The point: Exceptional achievement often comes packaged with difficult personalities.

Taking BS to mean behavioural science, Rory asks whether everything is BS. Not quite, but most things involve a heady dose of it, and many persistent problems could be solved faster by adding a psychological dimension. The converse does not hold: BS is not everything. It is necessary but not sufficient, and we should not assume every problem must be solved exclusively by behavioural science. Behavioural science without creativity, without a little whiff of the other kind of BS, may be sub-optimal.

“Everything is BS, but BS isn't everything.”

The point: Psychology should expand the solution space alongside other disciplines, not replace them.

The Kaufmann family asked Frank Lloyd Wright for a house next to their favourite waterfall. He was, Rory believes, past the deadline when at the last minute he asked: why does the house need to be next to the waterfall? It could be over the waterfall. The result was Fallingwater, an extreme case of re-asking the question.

“Why does the house need to be next to the waterfall? It could be over the waterfall.”

The point: Challenging the brief's assumption can produce a radically better solution.

Many people in advertising are slightly off-beam: Scottish, or posh Welsh like Rory. Scots are disproportionately good at it because they are close enough to the mainstream to understand it but can detach themselves from it. Gillian Tett calls this anthropological skill anthro-vision. Just as fish don't understand water, conventionally raised middle-class people cannot see their own milieu. Rory never realised he was Welsh until he left Wales, and at English school speech days finds it all too Anglo-Saxon.

“Just as fish don't understand water, middle-class people who've been entirely conventionally brought up can't see the water.”

The point: Being slightly outside a culture gives the detachment needed to see and use its rules.

Metrics of success should not be entirely fixed. If you set out to succeed one way and instead succeed massively in an unanticipated way, that is not a failure. If you go out to buy milk, forget the milk but buy a lottery ticket and win half a million pounds, that is not a bad day. The finance department, however, would say: sorry, your only metric is milk acquisition, so we must mark this a total failure.

“If you go out to buy milk, you forget to buy milk, but you buy a lottery ticket and win half a million quid, that's not a bad day.”

The point: Rigid success metrics misclassify unanticipated successes as failures.

Rory asks whether there is any country you would rather live in gastronomically than the UK, and doubts there is. Because Britain lacked a dominant food culture, it borrowed from everywhere, which makes for a more interesting food ecosystem. France is wonderful for the first couple of days, but by day three you find yourself asking whether you could just have a curry.

“France is wonderful for days one or two, but on day three you go, can I just have a curry?”

The point: A weak native tradition can be an advantage because it leaves room to absorb ideas from everywhere.

Optimising one or two dimensions does not necessarily resolve a contradiction. Another TRIZ-style move is to go to two extremes rather than one average. Don't produce a mid-priced vacuum cleaner: either produce an insanely cheap one by changing the rules or, like Dyson, an insanely expensive one.

“Don't produce a mid-priced vacuum cleaner.”

The point: The middle of a market is often the weakest position; polarised offers resolve trade-offs.

Rory, who admits he is no mathematician, sees a similarity between really good mathematics, such as Ole Peters' work, and advertising creativity. Solving a problem is not about being brilliant at maths but about knowing which kind of maths to use. A problem that seems almost impossible can become very easy if you reframe it. Facts are rarely universal truths; they depend on how data is collected, as snapshots or as time series.

“It's not about being brilliant at maths. It's about knowing which kind of maths to use.”

The point: Reframing a problem is often more powerful than applying more effort within the original frame.

Ask a single person what they enjoy and they will say disposable income, late nights, parties and exotic holidays. By that measure, having children is a really stupid idea. Except that once you have kids, your preferences change so that picking them up from school is more rewarding than an evening at Studio 54. We are not the same person after we have done some things, so asking people whether they, as they are now, want something narrows the solution set dangerously.

“Many, many things we do, we're not the same person after we've done them.”

The point: Testing ideas only against current preferences reduces exposure to upside discovery.

Following John Ralston Saul's argument that the West has given reason an exaggerated supremacy over intuition and creativity, Rory adds humour to the list of neglected human faculties. Humour is a problem-solving mechanism. Anyone with children notices how quickly their sense of humour develops: you can make a small child laugh before it can speak, which suggests something evolutionarily important is going on that deserves more investigation.

“You can actually make a small child laugh before they can speak.”

The point: Non-rational faculties like humour and intuition are evolved problem-solving tools, not distractions from reason.

Rory says he and his team tend to think backwards, and humour helps. Humour may be an evolutionary necessity: if you think backwards, 5 per cent of the time you are a genius and the other 95 per cent you are irritating. The benign violation lets you advance an opposing frame of reference to a group without it threatening their confirmation bias, a form of correction without threat. He worries about taking comedy off air, like a 1970s Fawlty Towers episode that was already satirising prejudice, and about a cult of humourless signalling of intensity.

“Five per cent of the time you're a genius; the other 95 per cent of the time you're kind of irritating.”

The point: Humour lets new or contrary ideas into a group without triggering defensiveness.

Rory finds AI models extraordinary and fascinating, but says he is not the Emperor Augustus: never in the last 30 years has he sat down and thought that what he really wanted was a scribe. What would make his life easier is something more like a helpful mate to chat things through with. The technology has been shaped by the urge to impress peers rather than by asking what people actually need.

“I'm not the Emperor Augustus. Never in the last 30 years have I ever sat down and thought, what I really want is a scribe.”

The point: Capability is not usefulness; AI products need to start from what humans actually want help with.

When LBC's Iain Dale stood as Conservative candidate for Tunbridge Wells, it emerged he had said negative things about the town. Rory's creative solution: swap constituencies and stand in Tonbridge next door, where making hatred of Tunbridge Wells the central plank of your campaign would win a landslide. There is a huge creative opportunity cost in trying to be right; often the way to solve a problem is to be interestingly less wrong.

“Quite often the way to solve a problem is just to be interestingly less wrong.”

The point: Reframing the context can turn a liability into an asset.

Responding to the old military line that stupid, energetic people are the most dangerous while clever, lazy people rank highly, Rory says laziness is not disconnected from creativity. The lazy person would rather sit and think of an easier way to do something than plough on and solve the problem by brute force, so there is probably some correlation between idleness and ingenuity.

“There probably is some sort of correlation in people between idleness and ingenuity.”

The point: Reluctance to use brute force pushes people towards cleverer, cheaper solutions.

Rory, who owns no denim, imagines pitching it: an impractical heavy workwear fabric, hot and sweaty in summer, fades badly, hard to maintain, popular mainly with indigent Californian workmen, and we predict it will be the great fabric of the 20th century. A Soviet system would never produce denim because it has no way to reward things whose popularity nobody can explain. Or pitch Wikipedia to investors, ending: and the best thing of all is people are going to write the whole thing for free.

“And the best thing of all is people are going to write the whole thing for free.”

The point: Hugely successful things often fail any upfront logical business case.

Purists criticise behavioural science for spawning airport literature, Rory's own included. He disagrees: the second-best thing we can do with behavioural science is practise it, the best is to disseminate it. Even if a slight misunderstanding of the literature leads someone to attempt something they otherwise wouldn't have tried, under the right testing conditions the effect is probably net positive, because it broadens people's conception of what value might mean.

“The second best thing we can do with behavioural science is practise it. The most valuable thing we can do is disseminate it.”

The point: Popularisation that prompts more experimentation does more good than purity protects.

In Deal on the Kent coast, the commemorative benches along the seafront could seat more than the town's population, while London Bridge station, which cost about 750 million pounds, has nowhere to sit. Rory spotted the feedback loop: benches say 'in memory of Uncle Bob, who loved sitting in this spot', which means there was already somewhere to sit. Places with no seating never get loved, so never get benches. His children have instructions to put a bench on platform 5 at London Bridge reading 'in memory of Rory Sutherland, who could never find anywhere to sit here'.

“In memory of Rory Sutherland, who could never find anywhere to sit here.”

The point: You can only solve problems once you spot the asymmetries that a system's own feedback loops create.

Rory loves the image of A. A. Gill, famously dyslexic, dictating his immortal prose by phone to The Times copy-takers, as journalists did before laptops. At the end of a paragraph the copy-taker would ask: 'Is there much more of this?' You can be that good and still, as with no man being a hero to his valet, have the functionary going 'yeah, okay'. He also cites Gill's book Pour Me: never call an alcoholic or addict lazy, because maintaining the habit takes extraordinary work and ingenuity.

“Is there much more of this?”

The point: Brilliance is invisible to those processing it as routine work.

At an Ogilvy training course in South Africa with Neil French in the early 2000s, creatives had to advertise for the tourist board when crime was through the roof. The advice was don't mention crime: reassurance only creates fear. A young copywriter from Chicago ran straight at it. He found a super-luxurious rail hotel, like the Orient Express, crossing the Karoo, showed its lavish interior and wrote: it's the carjacking capital of the world, so take the train. Rory and French instantly recognised him as one of their tribe.

“It's the carjacking capital of the world, so take the train.”

The point: Confronting the awkward fact head on can turn it into the persuasive reason to act.

Willy Brandt supposedly said God gave the Germans every virtue except a sense of proportion. Rory is a fan of losing it. Numbers dangerously suggest that all things have a linear relationship, but anything involving human psychology probably does not. Proportionate thinking assumes big change needs big inputs; marketing's glory is that small interventions can have enormous effects. If boards view their business purely in numerical and financial terms, they miss a huge area where value resides and can be created or destroyed.

“It's time we lost our sense of proportion.”

The point: Abandon the assumption that effects are proportional to inputs; human behaviour rewards disproportionate, small interventions.

When Rory first met Daniel Kahneman in 2009, Kahneman said he had no hope of making human behaviour more rational. What he did hope was to change how people gossip, so they would say things like 'I'm going on holiday, but I wonder if it's sunk cost bias because I've already paid the deposit'. Rory's version: if we change the way people discuss problems, so that opening them up to other fields and variables is a positive rather than a disruption, we have made 90 per cent of the progress.

“If we can simply change how people gossip, we've won.”

The point: Better decisions come less from fixing individuals than from changing the shared language used to discuss choices.

Rory suggests real-time video, via Zoom or live streaming, perhaps with influencers, is an unexplored area in charity fundraising. The live nature of something changes how we process it psychologically. Imagine going in real time to a school damaged by bombing or an earthquake and saying, let's see how much we can give this man who wants to repair it, with everyone watching the money arrive together.

“The very live nature of something fundamentally changes psychologically how we process the information.”

The point: Immediacy and visible collective giving make an appeal feel concrete and urgent.

Rory says the most vital thing in an advertising agency is a culture where it is acceptable to fail or be silly. You need an environment in which people can make stupid suggestions and still get promoted, because that is the only way enough ideas get aired for some to be good.

“Creating a culture in which you can make stupid suggestions and still get promoted.”

The point: Idea generation depends on removing the career penalty for being wrong or silly.

Rory cites Paul Graham's one-page essay 'Maker's Schedule, Manager's Schedule': to a manager, a 20-minute meeting mid-afternoon is just 20 minutes, but to a maker it destroys the whole afternoon. Writing his book, Rory found that when he blocked out three days, the relationship between time and productivity was completely non-linear: day one he wrote nothing, day two a little, and day three he produced 8,000 words. Flexible working could let us re-partition time, perhaps making Wednesdays sociable and Fridays secluded.

“Day one you wrote nothing, day two you wrote a little, and day three you produced eight thousand words.”

The point: Creative output depends on long unbroken stretches of time, so small interruptions cost far more than they appear to.

If you are rich and want to give to charity, the logical answer is to give the money, and nobody will criticise you. Or you can announce that you will match donations up to $200,000. Now you have not given $200,000; you have given $400,000. The logical answer is fine, Rory says, but it is not magical.

“You haven't given $200,000. You've given $400,000.”

The point: Reframing how something is given can multiply its effect at no extra cost.

Though a classicist, Rory found about half his university friends were mathematicians, partly because they had the most free time: you can either do maths or you can't. He supports teaching maths to 18, and philosophy and decision-making too, but thinks maths is valuable for reasons that have little to do with numbers. Mathematicians solve problems by redefining, reframing or changing the frame of comparison; some geometry problems are insoluble until you draw two parallel lines, then obvious. Computers do the calculation; the problem-solving skills are priceless.

“Maths is more valuable for its encouragement of problem-solving skills than for its teaching of numerical skills.”

The point: The real transferable skill is reframing a problem until it becomes solvable.

WPP chief executive Mark Read wrote that if you had asked McKinsey to regenerate Manhattan's Lower West Side, they would have proposed a new metro line costing five billion. Someone else came up with the High Line, turning a derelict elevated railway into a park and walkway. Once you define it as a transport problem you go down transport metrics: speed, efficiency, capacity. There is no metric for loveliness, no SI unit for perception. Just make the area gorgeous and you solve it without transport investment.

“There isn't a metric for loveliness.”

The point: How you define a problem determines which metrics and solutions you can see.

A mathematical logician called David Griffiths turned up, nobody knows why, at a traffic modelling conference. As everyone discussed how to model traffic jams, he raised his hand and said it might be better to model the not-cars. Each car carries a lot of data; modelling the gaps is simpler, because there are fewer gaps than cars, and may reveal things about jams that looking at cars never would. Rory says that is what behavioural science is about: looking at the not-cars.

“When you are modelling a traffic jam it might be better to model the not-cars.”

The point: Looking at the negative space of a problem can reveal what the obvious framing hides.

There are two kinds of chef, Rory suspects, as there are two kinds of art director or copywriter: the Mozart and the Beethoven. Mozart was a natural talent who could compose almost on the fly; Beethoven had deep wells that needed more drilling. Shakespeare reputedly wrote without blotting a word. Neither type is more talented: talent is only worth judging by its output. Rossini, by contrast, got rich enough that he largely stopped bothering.

“It's only really worth judging talent by its output.”

The point: Different creative working styles can be equally valuable; judge by results.

Prince's Hot Chicken in Nashville began when the original owner, a womaniser and drunk, kept staying out late. His furious girlfriend left him a meal of insanely over-spiced chicken as revenge. He came home drunk and thought it was absolutely delicious. The whole restaurant, and the Nashville hot chicken style, was built on a dish created as punishment, because when you are drunk that is exactly what you want.

“The entire premise of the restaurant was built around this thing which was created as a revenge dish.”

The point: Accidents and unintended uses can produce successful products no planner would design.

Playing the reverse benchmarking game with hotels, Rory notes that he has stayed in many, swanky and cheap, and fans of Moxy and CitizenM, yet not one has ever offered, even for money, a monitor in the room. He spends more time working in his room than watching TV, and being old and long-sighted would love a 34-inch monitor with a USB-C cable. Nobody has done it simply because nobody has thought of it.

“All it is is that nobody's thought of it.”

The point: Obvious unmet needs persist because whole industries copy each other's assumptions.

Norway's state broadcaster, on its second channel, experimented early with slow TV, including about 24 hours devoted to wood: much of it showing wood burning, then people chopping and stacking it. It caused a scandal, because whether you stack firewood bark up or bark down is a polarising, Marmite issue in Norway. Rory loves it as a broadcaster questioning the one thing everyone treated as a given, the time length. His brother-in-law, a BBC detective drama writer, says few writers can fill two hours without ad breaks.

“Whether you store firewood bark up or bark down is one of those weird polarising Marmite things in Norway.”

The point: Question the format constraints everyone takes for granted.

Rory's advice in one admittedly awkward sentence: marketers should look for things that are objectively similar but subjectively different, where the same thing can be made ten times cooler by presenting it differently, or subjectively similar but objectively different, where a cheaper thing achieves the same effect as an expensive one. He ends the talk on it: that gap is where to look for 10x gains.

“Look for things that are either objectively similar but subjectively different, or subjectively similar but objectively different.”

The point: The cheapest big wins live in the gap between objective reality and subjective experience.

Rory's favourite book, readable in about five minutes, is Robert Updegraff's Obvious Adams from 1916, one of David Ogilvy's favourites too. It seems hokey but is profound: an advertising man solves business problems simply by noticing things that seem obvious once pointed out. They stay hidden because business's need to maintain the pretence of rational, sequential thought closes off many possible avenues.

“Things which are seemingly obvious in retrospect once you point them out, but which are completely hidden from people.”

The point: Insight often comes from plain observation that formal reasoning processes filter out.

Rory blames Meta, Google and consulting firms, comfortable with and enriched by efficiency optimisation, for distorting marketing into chasing 3% improvements at the bottom of the funnel rather than billion-dollar discoveries such as the Pumpkin Spice Latte or Share a Coke. David Ogilvy said he had only five big ideas in his life, so presumably the finance department should have been furious with him for 30 of his 35 years. There is no proportionality between effort and value: marketing is treasure hunting, not mining.

“Trying to make it look like iron mining when really what you're doing is treasure hunting is a perversion of what marketing should be.”

The point: Creative value comes in rare big discoveries, so judging it by steady effort and output misreads it.

There are more good ideas we can post-rationalise than good ideas we can pre-rationalise. For years Rory travelled with about 15 plug adapters, several for each country, to charge two phones and a laptop. It took him ten years to realise that one universal adapter plus a three-gang extension cable caters for everywhere, and you don't have to unplug the hotel lamp or television to charge your laptop. Obvious in retrospect.

“There are just more good ideas that we can post-rationalise than good ideas we can pre-rationalise.”

The point: Many good ideas are only obvious after the fact, so insisting on prior justification loses them.

Rory cites a Silicon Valley investor who will only invest in something if it has an element of stupidity, some component that seems a bit ridiculous. From TikTok to Red Bull, and in government policy too, breakthrough innovations often contain something slightly ludicrous. The fact that they seem ludicrous explains why nobody has tried them before. Also, the logical ideas have all been tried.

“The fact that they seem slightly ludicrous explains why no one's tried them before.”

The point: Apparent absurdity is evidence that an idea is unexplored territory.

Rory calls 'why did the chicken cross the road?' the oldest question in behavioural science, and notes it contains two hidden assumptions: that there is only one chicken and only one reason. Every time we ask why, we should consider that there is more than one reason, more than one chicken, and that the chickens may not all have the same reason. Do that and you escape the tyranny of averages, and new sunlit uplands of creative possibility emerge.

“There's more than one reason and there's more than one chicken, and maybe the chickens don't all have the same reason.”

The point: Assuming a single cause shared by everyone hides the varied motives that offer new solutions.

Rory borrows an architectural idea from a book he had just downloaded: don't lay paths across a grass court in advance. Wait to see where people naturally walk, then put concrete where the muddy tracks have formed. Designing paths up front relies on assumptions about where people want to go that you cannot know. Likewise, if choices feel instinctively comfortable, lots of people make them; if they feel contrary to instinct, people find them off-putting.

“Paving the cow paths.”

The point: Observe what people instinctively do and design around it, rather than imposing assumed behaviour.

Borrowing the language of sex therapy, Rory says we have a problem with premature rationalisation: people leap to reason too soon. He loves Donald MacKinnon's description of the most creative scientists and architects as like sedate businessmen: not wacky, undisciplined artists, but people who were more relaxed and less impatient. Their relationship to time was fundamentally different.

“To borrow the language of sex therapy, we've got a problem with premature rationalisation.”

The point: Rushing to logical evaluation kills ideas before they form; creativity needs patience more than eccentricity.

For a man who loses his house keys monthly, Rory's answer is a design solution: a key rack right inside the front door. You have to spot exactly the right moment to intervene in a behaviour. Since the keys are already in his hand as he comes through the door, hanging them on the wall needs far less behavioural change than remembering to put them somewhere in the bedroom. Keys are a System 1 thing, habitual and unthinking. If he loses them out of the house, he should stop drinking.

“You have to spot exactly the right moment at which to intervene to adopt a new behaviour.”

The point: Behaviour change works best when the new action sits at the moment the old one already happens.

Rory calls it the greatest recontextualisation in political history. In 1984 Reagan, at 73, had fumbled the first debate and advisers urged him to tackle the age issue head-on with rational argument. He told them not to worry. When the moderator asked if he could function under Cuban-missile-crisis pressure, he said he would not make age an issue: 'I am not going to exploit, for political purposes, my opponent's youth and inexperience.' One joke and the whole issue was killed. Mondale later said that was when he knew he'd lost. You can communicate things with humour you can't say straight.

“You can communicate things with humour that you can't say straight.”

The point: Reframing a weakness with wit can neutralise it far better than rebutting it.

Rory loves what Will Guidara, in Unreasonable Hospitality, calls reverse benchmarking: instead of asking what your best competitors do well to copy, ask what they do slightly disappointingly that you can hit out of the park. For Guidara it was coffee and beer. Breakthroughs often do what everyone else does not quite as well, and pour resources into what they neglected. Diners offered huge choice: over easy, sunny side up, substitutions. McDonald's did the opposite, cutting choice and delivering speed.

“You ask what do they do slightly disappointingly that we can hit out of the park.”

The point: Win by excelling at what competitors neglect, not by matching their strengths.

Most of what we do rests on tacit knowledge, things we know without knowing them. On Clarkson's Farm, Kaleb says of reversing a trailer through a narrow gate: I can't tell you how to do it, you just have to know. An Imperial College engineering professor would mess it up; someone who learned unconsciously does it first time. At Cambridge, Rory's college faced Marks and Spencer, which paid to reverse articulated lorries into a tiny entrance, and watching the drivers was art. Restaurateurs similarly know, without explaining, what makes a café a café.

“You can't explain how to do it, you've just got to know.”

The point: Much expertise is unconscious and cannot be captured by explicit theory.

Rory thinks it a pity that advertising confined creativity to making communications rather than defining problems. Look at problems through a behavioural, perceptual lens and the solution space gets much bigger. Francis Crick said that by blundering around they stumbled on gold, but they were looking for gold. Similarly about 20 percent of bees fly around at random, but they are looking for nectar, and when a rogue bee finds a new source it shares it with the hive. Rory's plea to the audience: if you find something, share it.

“By blundering around we stumbled on gold, but the fact remains we were looking for gold.”

The point: Purposeful randomness, plus sharing what it finds, is how new sources of value are discovered.

Alongside the better-known rules, the book opens with several more: don't design for average; the nature of our attention affects the nature of our experience; the problem with logic is that it kills off magic; a good guess which stands up to observation is still science, and so is a lucky accident; dare to be trivial; and, finally, if there were a logical answer, we would have found it. Together they license looking for cheap, odd, non-obvious fixes to stubborn problems.

“If there were a logical answer, we would have found it.”

The point: Problems that have resisted logical solutions probably need illogical ones, so permission to be trivial, lucky or odd is valuable.

This is why comedy matters: it's easier to start silly and reel it back than to start boring and push it forward. Rory's hypothetical: keep the existing tracks and spend the money recreating the world's greatest steam locomotives, so your 9:15 from Manchester is pulled by a Union Pacific Big Boy or a replica of the Mallard. It would be slower, but every tourist in London would go to Manchester just to ride the train. He also floats 'Train Twister' as a cure for overcrowding: turn proximity to strangers from annoyance into entertainment.

“It's much more easy to start silly and reel it back than it is to start boring and reel it forwards.”

The point: Deliberately absurd ideas widen the solution space and can be refined into practical ones.

Discussing why Chinese food is underrated in Britain, Rory accepts it became slightly naff through ubiquity. The landlord explains that huge menus full of unfamiliar regional dishes make it hard for newcomers; Rory sums up that the choice architecture is terrible, built for people who already know. He suggests there is probably scope for someone to do for Chinese food what Dishoom did for Indian: reinvent the experience of a familiar cuisine.

“The choice architecture is a total mess, isn't it?”

The point: A category made naff by ubiquity and confusing choice is ripe for reinvention through presentation.

If you aggregate data you're probably using it for self-justification; if you look for anomalies, like a Simpson's paradox or outliers, you're using it for illumination. Cops are better than scientists because they accept an investigative phase that doesn't demand evidential value. Son of Sam was caught because someone mentioned seeing a traffic warden unusually late. It was an outside chance, but officers checked who got a ticket that night, found his car with guns visible in the footwell, and called for backup. Science publishes only the evidential phase; the 'Night Science' podcast argues the noticing matters more.

“If you're looking for anomalies, you're probably using it for illumination.”

The point: Discovery depends on chasing small anomalies that would never count as evidence.

Responding to Hemingway's line that the first draft of everything is rubbish, Rory says the world is designed around rational arguments: if you can only benefit from things argued for in advance, you get a very limited solution space. Even in science, most progress is unintended or observational, what Taleb calls stochastic tinkering: rather than designing the perfect steam engine on paper, you build a pretty good one and keep fiddling to improve it. Practice informs theory. And we promote people for winning arguments, which is a poor proxy for solving problems.

“Even in science, probably the greater part of progress is unintended.”

The point: Iterating in practice beats insisting on a fully reasoned plan before acting.

One of the most important things to do with AI, says Rory, who is part Welsh, is use voice dictation and pretend you're Welsh. With search it pays to be concise: 'cheap toasters'. With AI you want to be extra verbose: I'm looking for a toaster, I'd like it brown and green, it must fit under my units, and go on a bit longer than strictly necessary. Dictating, you naturally do that, and you get far more inspired results because AI combines requirements in parallel rather than sequentially.

“Use voice dictation, pretend you're Welsh.”

The point: Rich, rambling context gets better results from AI than terse search-style queries.

Talking about Greg Jackson of Octopus Energy, Rory says that if you put ten hardcore systems engineers in a room with ten marketers, with enough drink flowing that they build a rapport and respect each other, half the world's problems would be solved quickly. Many problems are a weird mix of engineering and psychology. He adds that marketers err by justifying themselves in the language of finance; the real value of marketing is philosophical, a new outlook that opens up new ways of solving problems.

“The value of marketing is not what we do, it is how we think.”

The point: Hard problems usually need engineering and psychology together, and marketing's worth is its way of seeing.

Asked how he got good at this, Rory says his one trick is probably ADHD: he is not interested in the middle of things, only in deep philosophical questions and in trivia. In complex systems you can either tinker with details and get surprisingly valuable responses, or invent a new paradigm. It mirrors the direct response rule from Claude Hopkins's era: test either a single word or a completely different advertisement. Test five things at once and you cannot tell which one made the difference.

“You either test one thing or you test everything.”

The point: The big payoffs lie at the extremes: tiny details or whole new paradigms, not the middle.

Rory predicts a successful agency in 2035 will spend 30 per cent of its time answering briefs and 70 per cent proactively finding problems to solve, because AI makes execution cheap. Picasso painted his own paintings and sold them; the Beatles did not wait for someone to request a song about a submarine. Commissions make sense for Fabergé eggs, where gold and craftsmanship are costly, but Wedgwood made plates from cheap clay at scale. Cannes should stop being a retrospective awards festival and become a trade show: 'that fried chicken ad for Latvia? $50,000.'

“The Beatles didn't wait for someone to come along and request a song about a submarine.”

The point: When production becomes cheap, creators should make work speculatively and find buyers, rather than wait for commissions.

The best review Rory got for Alchemy, in the Wall Street Journal, came from a coder who said the book made him realise he had two strings to his bow, not one. He used to think he had to solve every problem with code and brute force; now he saw he could also solve problems with psychology, and sometimes that was easier.

“I now realise I've got two strings to my bow, not one.”

The point: Psychology is a complementary problem-solving tool for technical people, sometimes the easier one.

Advertising works more like detective work than science: act on a hunch that tells you where to look. Investigating the Yorkshire Ripper, police fixated on a hoax tape's Wearside accent. Two cops interviewing men seen in red-light districts broke the ice with wives: 'Now's your chance to put your husband away for good.' About 90 per cent laughed and 10 per cent got angry. Only the Sutcliffes did neither. With his tooth gap, lorry driving and footwear matching prints, they flagged him; their boss asked if he had a Geordie accent, said no, and bollocked them. Similarly, Levi Bellfield was caught after an ex mentioned defaced magazine photos.

“Every single person they played that joke on, about 10% of them got angry and 90% laughed.”

The point: Anomalies and hunches deserve investigation even when they lack evidential weight; fixed hypotheses blind institutions.

What taught Rory the importance of the overnight test, and of going away to distract yourself, is cryptic crosswords (apologising to Americans, who barely have them). You can stare at a clue completely bemused, then come back to it after a period of self-distraction and find the answer absolutely self-evident.

The point: Stepping away lets the unconscious solve what concentrated effort cannot.

Speaking to a drugs company, Rory heard that some compounds tested for autoimmune diseases like rheumatoid arthritis were appalling: they boosted the immune system, the last thing you want when it's attacking the body. They were shelved. Then someone realised that for cancer immunotherapy, a drug that strengthens the immune system is not a bug but a feature. Most decision-making is top-down, but progress is bottom-up or horizontal, and modern problems increasingly sit between specialisms, needing people who know a bit about three things.

“That's not a bug, that's a feature.”

The point: Failures in one context can be solutions in another, found by people who span specialisms.

Code-breaking is very hard if you do not know what you are looking for, but much easier if messages end in 'Heil Hitler' or start with a girlfriend's name. Rory recounts a very valuable German stationed somewhere in the North African desert whom Bletchley was desperate should not be attacked. Every single day he posted exactly the same thing: nothing to report. Knowing exactly what he was saying, they only had to find the dial settings that produced it.

“Every single day he would post exactly the same thing, which was nothing to report.”

The point: Predictable routines leak information; small regularities are the key to breakthroughs.

Businesses pour energy into measuring and quantifying, assembling statistics for presentations upwards and to shareholders, but put little effort into asking what the causes might be. Rory calls this the hyperactive what and the lazy why. Once a satisfactory rational explanation appears, the brain says 'got that, move on, nothing to see here'. Yet what questions rarely generate new ideas, whereas an interesting why gets you 75 per cent of the way to a solution. Whats provide creative constraints; whys provide creative springboards. Businesses still spend 90 per cent of their time asking what.

“Whats generally provide creative constraints and whys generally create creative springboards.”

The point: Finding a new explanation for behaviour opens whole new solution spaces that more data never will.

Rory recalls that in the 1980s it was common to ridicule hybrid degrees such as theology and film studies. He always thought the opposite: those hybrid degrees are brilliant, and one of his best colleagues studied theology and philosophy. Borrowing E. O. Wilson's idea of consilience, he argues the interesting stuff in science often happens at the intersections between disciplines, as with behavioural economics, rather than in their interiors.

“The interesting stuff in science often happens at the intersections, as with behavioural economics.”

The point: Combining disciplines yields insights that specialists inside a single field miss.

For the curious 20 per cent of the population, far more things are interesting than anyone anticipated. On YouTube, Ask a Mortician describes the aftermath of the Kennedy assassination from a funeral director's view: they had to buy a coffin unexpectedly, choose a funeral home, and the agents who commandeered the hearse to take the body to Love Field had no idea how to detach the coffin. The most over-studied hours of 20th-century history gain a whole new dimension from a slightly weird point of view. James Dyson's advice applies: view the problem the wrong way round.

“A whole extra dimension appears just through adopting this slightly weird point of view.”

The point: Looking at a familiar subject from an unusual vantage point reveals what everyone else misses.

Asked what matters most when launching a product, Rory says the laws of conventional economics and physics can't be rewritten: you can't do anything about gravity, so confining yourself to them leaves little room for creativity. The laws of psychology, by contrast, are gloriously mutable and gameable. Look for a psychological edge or arbitrage and you're far more likely to have a breakthrough than by trying to be cheaper or faster. Have a great idea for an electricity company, but make sure you've got a pink octopus.

“The laws of psychology are gloriously mutable, gameable.”

The point: Distinctiveness is easier to win through psychology than through price or speed.

Rory keeps a one-kilowatt-hour portable battery and solar panels in his car. Citing Helen Edwards's advice to look at extreme use cases to understand innovation, he describes a lorry driver who uses such a battery with a Starlink Mini on the cab roof, getting fast internet and running an air fryer at every stop. Rory realised picnics would never be the same: you could bring an espresso machine. He wonders why carmakers have not exploited the 120 kilowatt-hours under an electric motor home, enough to live off-grid for weeks.

The point: Extreme users reveal uses for a technology that mainstream customers have not yet imagined.

A Royal Society of Arts paper on design thinking describes a "missing first diamond": a period of comfortable messiness at the start of every brief, asking what is really going on in an open-minded, exploratory way, before assuming which tool will solve the problem. Even creative businesses skip it. A colleague pointed out that behavioural science practices struggle to win business precisely because they operate in that diamond. Put bluntly, no one has a budget for solving a problem they did not know they had.

“No one has a budget for solving a problem they didn't know they had.”

The point: Diagnosis is undervalued because budgets are allocated to known problems and predetermined solutions.

Two people at Unilever were briefed to invent instant soup and succeeded with a powder you add boiling water to. One declared the job done. The other, a Dutchman, felt only half the problem was solved, because you still had to wait for the kettle to boil. Although it was off brief, he became obsessed with producing boiling water instantly and ended up creating Quooker. Rory admits the puritan in him balks at paying nearly four figures for a fast kettle, yet he would never go back and has found new uses for instant boiling water.

The point: Pursuing the unsolved half of a problem, even off brief, can create a whole new category.

Tina Seelig, who teaches entrepreneurship at Stanford, gave students five dollars in an envelope: once opened, make as much money as you can in two hours. Almost invariably the biggest earners ignored the five dollars as an irrelevant anchor. The winning team made 650 dollars by realising their real asset was a ten-minute presentation slot before the graduating entrepreneurship class, and selling it as advertising to a headhunting firm. They walked on stage, said 'we made 650 dollars, here's an ad', and left.

“If you start with a financial frame, your solution space becomes entirely constrained.”

The point: The resource you are handed can anchor you away from the most valuable asset you have.

Rory's trifecta, or sweet spot, says that whenever you try to change behaviour, a business or an economy for the better, you should give weight to three areas: technology, economics and psychology. The most interesting progress, and many of the most interesting businesses, arise where someone has all three in alignment. Things go wrong when a dominant engineering or economics culture ignores the psychological leg.

“The really interesting stuff doesn't actually happen within disciplines, it mostly happens at the intersection between them.”

The point: Solutions that combine technology, economics and psychology beat those that optimise only one.

Sage once gave Rory a dual-boiler coffee machine they did not need to, so his reciprocation bias kicked in and when replacing his toaster he looked at Sage. One model has a button labelled a bit more. It serves two purposes: a crumpet not quite brown enough can get a little extra without being cremated, and toast that went cold while you wandered off can be revived. He calls such features Easter eggs, and thinks product development should be full of them.

“One of them has a button entitled just a little bit more.”

The point: Small delightful features, Easter eggs, create disproportionate affection for a product.

Nobody asked what a scroll-work ice cream would look like. A faulty extruding machine on a wobbly ice-cream conveyor belt produced vanilla ice cream in a shape far more attractive than a slab, and someone noticed: the Wall's Viennetta. Penicillin, Viagra and the Viennetta were not found by defining the question first. Rory's definition of creativity: an answer to a question nobody has yet got round to asking. Most creative work mixes luck with inspired observation.

“Creativity is an answer to a question that no one's actually got round to asking.”

The point: Great products often come from noticing accidents, with the rationale supplied afterwards.

Rory always asks of a problem: are we looking at the wrong things? Are these things true or just assumptions? And just because something happened in the past, does that mean it will continue? He cites Danny Dorling's book Slowdown: UK car use was already declining before the pandemic, and the amount of stuff consumed in the UK fell around 2006 or 2007 with no loss of economic growth. We project forward as if the world were linear.

“Are these things true or are they just assumptions?”

The point: Questioning focus, assumptions and extrapolation exposes hidden errors in forecasts and plans.

Rory has become interested in TRIZ, a Soviet-era innovation method devised by a patent clerk in the 1950s who was rewarded with five years in a labour camp. It treats most problems as contradictions. You want an umbrella big enough to keep off the rain, but a big umbrella in the wind wrenches you off your feet. The gust-proof umbrella resolves it with two overlapping canopies so air can escape. It makes a loud farting noise, but that beats losing the umbrella altogether.

“It does mean your umbrella makes a loud farting noise, but that's probably better than losing the thing altogether.”

The point: Innovation often comes from resolving an apparent contradiction rather than optimising one dimension.

Rory says he learned none of this at university; he learned it all at Ogilvy. University teaches you to reduce every problem to a two-body problem that can be solved mathematically to a single optimal answer: pretend the problem is that simple, solve the pretend model, pat yourself on the back. Only at Ogilvy did he learn that the trick is to find a completely different variable nobody is looking at, and play around with that instead.

“Pretend the problem is that simple, solve for the pretend simplistic model and then pat yourself on the back.”

The point: Breakthroughs come from adding neglected variables, not from optimising a simplified model.

Rory's Mad Masters course does not tell people how to do their jobs: 90% of the time, do the logical thing. He cites a phrase attributed to Whitehead that bursts of thought are not to be deployed all the time but like cavalry charges in battle, at judicious, decisive moments. The same goes for behavioural science and creativity: try to be creative about everything and you go insane. Some things must run on habit and best practice, but leave space to ask 'what if the opposite were true?'

“They are to be deployed rather like cavalry charges in battle, at judicious, decisive moments.”

The point: Save radical rethinking for decisive moments and run the rest on habit.

No process can reliably produce an idea like the meerkat. All you can do is maximise randomness and then maximise your ability to spot significance within it, which is also how most science works: nobody was intending penicillin, and Viagra was meant to cure angina. You need a framework of order, but the complete elimination of randomness is really dangerous in the medium to long term, because that's where the really valuable stuff comes from.

“The complete elimination of randomness is really, really dangerous in the medium to long term, because that's where the really valuable stuff comes from.”

The point: Breakthroughs come from generating variety and recognising the lucky accident.

Lidl and Aldi's middle aisle, where you go in for milk and leave with a wetsuit and a paddleboard, reveals a philosophical truth. A successful life or business needs two modes. Walmart shopping: know what you want and pursue it efficiently. TJ Maxx life: swan around hoping to get lucky. Friends complain they went to TJ Maxx for a yellow cardigan and found nothing; you idiot, you go in like a shark. Taleb calls it increasing your exposure to positive optionality. That's also why fame pays: strangers bring you business.

“You don't go into TJ Maxx looking for something specific. You go in like a shark.”

The point: Balance efficient pursuit of known goals with undirected exploration that creates chances to get lucky.

Rory endorses Matthew Syed's point that society suffers not from a lack of intelligence but from a singularity of viewpoint: all the intelligence is directed at the same focus. Economics courses, business schools, MBAs and McKinsey compound the problem by having lots of highly intelligent people looking at exactly the same thing. Globalisation means more people see the world the same way, for too long.

The point: Diversity of perspective matters more than raw intelligence for solving problems.

Rory ends with a question: we assume our interactions with AI should be instantaneous, but are we sure? It might be more useful to ask an AI to send you ideas for holidays in Greece over the next three or four months, so you can see options, refine them, mull them over and discuss them, rather than decide immediately. Along with Hartmut Rosa, he thinks there are things we should deliberately slow down, for our sanity and our productivity.

“What does slow AI look like?”

The point: Optimisation assumes faster is always better, but some decisions improve with deliberate slowness.

Creativity and logic are not opposites but two strings to your bow, used in parallel. One of the most important questions in a police investigation or in science is: what if we are proceeding on an incorrect assumption? What if this person didn't do it? What if the Yorkshire Ripper doesn't have a Geordie accent? What if, as some statisticians argue in a current UK case of a nurse convicted of murders on a neonatal ward, no crimes were committed at all? What if is a creative question, what Charles Sanders Peirce called abductive inference.

“The question what if is fundamentally a creative question.”

The point: Deliberately questioning the governing assumption is a creative act that logic alone does not perform.

The best advice Rory received came from a university friend, a mature student who had organised the Isle of Wight festivals: when everyone else zigs, zag. BBH later made it its slogan, and everyone thought it invented it. We are a herd species, but big gains never come from pursuing the same metrics of success as everyone else. Define success for yourself, or you will judge it by comparison with your school and university peers. Going further back, Montaigne: your own happiness is too important to outsource to someone else's values.

“When everybody else zigs, zag.”

The point: Outsized gains, and personal contentment, come from refusing to adopt everyone else's measures of success.

Asked how he finds insights, Rory says it helps to have the mind of a naturalist. He cites Robert H. Frank's The Economic Naturalist, which asks students to explain anomalies, as Darwin did with finches' beaks or the peacock's tail. One of Frank's great questions: why do people buy a wedding dress they will wear once, while men rent tuxedos they might wear 50 or 60 times in a life? Rory admits he does not fully know the answer; the point is to hunt for such anomalies and work out what underlies them.

The point: Behavioural anomalies are clues to hidden motives and worth collecting like a naturalist.

Cognitive diversity brings surprises. There's a fine line between OCD and creativity, so how healthy is it that we are effectively medicating creativity? People slightly on the autism spectrum make very good social scientists, because what is instinctive to neurotypical people they must work out from first principles, giving them heightened rather than weakened understanding of social interaction. Advertising has many neurodiverse people and dyslexic creatives, and at one point, at the height of its success, most of Tesco's board were reportedly dyslexic.

“How healthy is it that we're effectively medicating creativity?”

The point: Apparent cognitive deficits can give people analytical advantages others lack.

Businesses fall in love with their models and start using them where they do not apply, which is a great competitive weakness. The way to win, Rory says, is to find out something your competitor assumes is true that isn't. Take something everybody assumes is true and pretend the opposite. That is the opportunity for innovation and experimentation.

“The way to win in business is to find out something your competitor assumes is true that isn't.”

The point: Shared false assumptions in a market are the richest source of advantage.

Rory calls the famously rude Wong Kei restaurant in London's Chinatown his perfect example of the principle that the opposite of a good idea can be another good idea: the opposite of where the mainstream market sits often provides an opportunity. Being so rude becomes the distinguishing feature, like theatre. But framing is essential: told in advance what to expect, it is one of the funniest experiences of your life; turn up expecting the Marriott and it is awful. His tip: ask for a knife and fork instead of chopsticks, which drives them practically insane.

“Usually the opposite of where the mainstream market is actually provides you with an opportunity.”

The point: Deliberately reversing a category norm can create a distinctive offer, if customers are primed to expect it.

Rory echoes the line that you can only find truth with logic if you have already found it without it. Many great discoveries come by accident, intuition, hunch or instinct; we try them, they succeed, and we post-rationalise them, but logic played no decisive part in finding them. He calls standard economic theory and literal market research the two broken binoculars of business decision-making.

“You can only find truth with logic if you've already found truth without it.”

The point: Logic validates discoveries after the fact; it rarely produces them.

Rory's hero Ludwig von Mises, of the Austrian School that treated value as entirely subjective, said there is no sensible distinction in a restaurant between the value created by the man who cooks the food and the man who sweeps the floor. One produces the primary good; the other creates the context, mood and mental state in which it can be enjoyed. Rory reads the cook as manufacturing and the sweeper as marketing. Serve Michelin-starred food in a restaurant that smells slightly of sewage and nobody will enjoy the meal. There are two ways to improve that restaurant: make the food even better, or get rid of the smell. Because we believe the objective thing is what matters, we keep improving the food.

“There's no sensible distinction to be made in a restaurant between the value created by the man who cooks the food and the man who sweeps the floor.”

The point: Context-creating work is part of a product's value, not overhead, and fixing the context can add more value than improving the product.

Variants: Sometimes told without von Mises as the bare sewage example, with the punchline 'No, get rid of the smell.' Other tellings add a bent fork tine or surly service, contrast the French physiocrats who thought only land created real value, or frame the meal as the food multiplied by the setting and expectation.

Most of the London Overground already existed as Silverlink Metro, branded like a mainline railway, and almost nobody used it; Rory once rode it from Stratford to Richmond as the only passenger. Londoners think the Tube map is a map of London: a British Rail logo means 'for leaving London', the roundel means 'for me'. Transport for London spent about £200 million on trains, lighting, stations and a few miles of track, then pretended it was a Tube line, coloured it orange and put it on the map. Usage rose about 400 per cent almost at once, and it now carries as many people as the £20 billion Elizabeth line. In 1989, wanting to live in Peckham because it was handy for Shoreditch would have got you committed.

“It's £20 billion worth of infrastructure that was mostly created with pixels and ink.”

The point: Making something mentally available can create as much value as building it, often for a fraction of the cost.

Variants: The value created varies by telling: £3 billion, £6 to £8 billion or £20 billion; the usage jump is fourfold, tripled on day one, or 400 per cent in a month. One version frames it as bottom-up thinking against Crossrail's top-down tunnel; another as the brain's inability to use two maps at once.

Everyone knows the sinking feeling when the plane stops a mile from the terminal: it's going to be a bus, and you feel cheated of your air bridge. Landing at Gatwick on easyJet, Rory heard the pilot announce bad news and good news. The bad news: a plane was blocking the gate, so no air bridge. The good news: the bus would take them all the way to passport control, so they would not have far to walk with their bags. That is always true of the bus, but nobody had ever pointed it out. He had rebranded the bus from an inconvenience into a conveyance, sparing a trudge past 20 Toblerone stands. Rory urges you to say it loudly next time: you synthesise happiness in everyone within earshot.

“The good news is that the bus will take you all the way to passport control.”

The point: Redirecting attention to a genuine but unmentioned benefit can turn a grievance into a positive at no cost.

Variants: One telling makes it a British Airways pilot and 'right next to immigration'. Rory sometimes adds that if a BMW 7 Series collected you from the plane you would think it marvellous, and draws the moral that marketing is multiplied value, not added value.

Range anxiety contains two words: range, a matter of physics and battery density, and anxiety, a matter of psychology. The world spends hundreds of billions on the first and almost nothing on the second, which is far cheaper to fix. Rory learnt how irrational it is driving his Lotus back from Wales: the gauge hit 16 per cent, went amber, and he was white-knuckled, though he still had 55 miles. His wife's electric Mini shows 55 miles as 55 per cent and they drive around without a care. So stop showing every 1 per cent drop: say '80 per cent plus', as old petrol gauges showed quarters. And make chargers visible: petrol stations evolved like peacocks, while chargers hide behind industrial estates.

“Maybe the solution isn't more lithium. It's more neon.”

The point: Before spending billions on a technological fix, ask whether the problem can be solved more cheaply in the head.

Variants: The figures shift (50, 55, 56 or 58 miles at 16 per cent). Rory adds that bigger batteries may feed the anxiety, since a 1,500-mile car would show 150 miles as 10 per cent, and that oversized batteries serve perhaps 2 per cent of journeys. Another fix: a CarPlay panel listing nearby rapid chargers.

Restaurants love wine because nobody knows what a bottle should cost: you cannot charge £40 for a Johnnie Walker, but a six-euro bottle sells for £30 or more. So they engineer the choice while you think you are choosing. Wine glasses are already on the table, removed with a slight huff if you decline. The drinks menu is called the wine list: pages of red and white, with a crappy laminated page at the back for the deviants and perverts who want something brewed or distilled. The sheer genius is that they bring one list, to one person. Wine is the only drink a table can share, so he asks 'red or white?', and it is game over for the gin drinkers.

“The second that guy says red or white, it's game over for the gin drinkers.”

The point: Props, naming and who holds the menu quietly steer a group to the most profitable option while everyone feels they chose freely.

Variants: In a telling for Diageo, Rory adds 'still or sparkling?', which makes asking for tap water awkward. One version ends 'game over for free will'.

How we perceive price, colour and taste is not objective. Painkillers are more effective when branded, and more effective still when people are told they are expensive. Nurofen sold pricier variants, for period pain or for cold and flu, that were in some cases chemically identical to the standard product, yet taking something labelled for period pain that costs a bit more will actually reduce period pain more. Painkillers work better if red; sleeping pills better if blue, except, apparently, in Italy, perhaps because blue is the colour of the Azzurri. Rory says calling this a hack is dubious: the painkiller genuinely becomes more effective. He complains you can't get expensive aspirin any more.

“I haven't got an 89p headache, I've got a £3.50 headache.”

The point: Expectations created by price, branding and colour change the real effect, so they are a legitimate source of value.

Variants: The headache price varies (49p, 79p or 89p against £3.50 or £3.90). He often pairs it with wine tasting better from a heavier bottle, and suspects the Italian explanation is itself a post-rationalisation.

Moxy, a Marriott brand, reverses the normal hotel. Rooms are small, in good locations, with no room service, no laundry, sometimes just hooks instead of a wardrobe, though the bed, TV and Wi-Fi are good. The investment goes into the ground floor: a 24-hour barista and bar where you check in, meet people, take video calls and hang around for hours after checkout, like a cooler WeWork; most hotels render you homeless at 11am. Moxy told Rory that the 90 per cent of guests who understand the trade-off love it, while the 10 per cent who turn up expecting a Marriott are furious. Rory, nearly 60, loves it for a night or two, though not for a week.

“You focus people on the one thing where you're amazing and then downplay everything else.”

The point: State the trade-off upfront and over-deliver on one salient dimension, and customers will happily accept being merely adequate elsewhere.

Variants: Some tellings stress expectation management (turn up expecting the Four Seasons and you will be livid); others stress the reallocation of investment, which a procurement mind reads as a deformed hotel. CitizenM is cited as the same model, and he once went down at 3am after eye surgery to ask for a flat white.

Perception is leaky: we think we can separate sight, sound, smell and taste, but we cannot. The strongest example Rory knows is that your car drives better after it has been valeted. It is not just cleaner: it seems quieter, smoother, accelerates better and corners more nicely, though nothing mechanical has changed. An engineer friend was so uncomfortable with this that he invented theories about cleaning tightening the body panels; complete rubbish, says Rory, it is entirely psychological. It is also why people always clean a car before selling it. So if you want to use the McGurk effect to save money, don't buy a new car: have your existing one valeted.

“The cheapest way to get a new car is to have your existing car valeted.”

The point: Improving one sensory cue lifts perceptions of unrelated qualities, so a cheap surface fix can feel like a product upgrade.

Variants: Often listed alongside wine tasting better from a heavier bottle and branded painkillers working better. He calls the result 'a placebo car'.

A television claims a billion colours but produces only three: red, green and blue, matched to the three types of cone in human eyes. Show yellow and there are no yellow photons; red and green pixels trigger the same cone response and your brain makes yellow. It is a biological hack: a TV reproducing the whole spectrum would cost a fortune. The box never says 'optimised for higher primates', but it is. Your dog can't understand why you paid so much, and a pigeon or parrot, which sees ultraviolet, thinks it is a washed-out piece of junk. Because perception, not physics, drives behaviour, designing for how people perceive is legitimate and far cheaper.

“Your dog can't understand why you paid so much for that television.”

The point: Design for the receptors people actually have: improvements outside human perception are wasted, and working on perception is legitimate.

Variants: Rory sometimes adds that magenta does not exist in physics, coins 'innervation' for synthesising value in the head, or tells of scientists baffled that parrots ignored photographs of ripe bananas because print is calibrated for human eyes. He says economics often produces 'ultraviolet light': things economists value that humans cannot see.

Professor Noriaki Kano of Tokyo split product attributes into three: threshold attributes, like a milk carton that doesn't leak, which nobody praises but whose failure loses you for ever; performance attributes, like sound quality, where more is linearly better; and excitement attributes, peripheral touches that delight out of all proportion. Buying a 1980s cassette deck, you should have judged build and sound. Instead everyone pressed eject. If the door opened with a damped, counterbalanced hiss, you had to have it; if it went clack, it was obviously a piece of junk. DVD players later let you open the tray by remote control, whose only possible purpose was to show off the mechanism.

“If you pressed eject and it just went clack, that was obviously a piece of crap.”

The point: Small, functionally irrelevant delighters can decide a purchase, yet they are the first thing a cost-cutter removes.

Variants: Rory links it to the DoubleTree cookie, arresting precisely because it is unnecessary, and to Chinese buyers who love the in-car gimmickry that European carmakers strip out.

Royal Mail spent a fortune improving efficiency and next-day reliability, then found no correlation between objective service in an area and how much people liked it. Some districts with superb delivery were indifferent; some with ropey service loved it. Alex Batchelor, then marketing director, tested a hunch: the leading determinant was whether people liked their postie. FedEx-level efficiency with a miserable postie and nobody liked Royal Mail; a sporadic service from a postie who said 'I knew you were on holiday, so I left it in the back porch' and it was the best organisation in the world. We have no evolved skill at judging postal logistics, but half a million years of judging whom to like and trust.

“A personal experience at a timely moment trumps anything you can do in advertising.”

The point: Find the proxy customers actually judge you by before spending on the metric the business thinks matters.

Variants: Rory's follow-ons: send the grumpiest 5 per cent of posties to charm school; move 10 to 20 per cent of marketing budget into the call centre and pay the best staff six figures; expect human discretion to grow more valuable as AI spreads.

If you want people to finish a course of antibiotics, and so avoid resistant bacteria, Rory says, don't give them 24 white pills. Give them 18 white pills and six blue ones, and tell them that when the white ones run out they should take the blue ones. The pills can be chemically identical, but more people will reach the end of the course, because a sequence that comes in stages, with a milestone in the middle, creates a sense of progress. It is the same psychology as a software loading bar: if something appears to be happening, we wait. It does not appear to make sense, which is part of its value. Rory offers it as his own, untested idea.

“Don't give them 24 white pills; give them 18 white pills and six blue ones.”

The point: Breaking a task into visible stages with a midpoint makes completion more likely, without anyone needing to know why.

Variants: One telling makes it 20 white and four blue. Gmail is cited as a version: it shows only unread mail, so you plough through 12, then find 23 more.

Joking that he is sponsored by the Welsh Tourist Board, Rory says that if you are English the best place to holiday is Wales, because you all think it is going to be a bit rubbish, and then it isn't. You expect not to find a latte or anything decent to eat, and discover a culinary mecca. In the unremarkable café at Cwmbran station, a Turkish-Welsh man turned out to be the god of panini making. According to Kahneman and other happiness researchers, what makes us happy is things slightly better than we expected. So optimise for pleasant surprises. A world where everything is perfect, like Singapore, where a train three minutes late triggers meltdown, leaves no room for them.

“All of you think it's going to be a bit shit and then it isn't.”

The point: Experience is judged against expectation, so low expectations exceeded produce more delight than high expectations met.

Variants: He adds: always rent a car on holiday, because it enables lucky accidents. In a pub telling, the landlord says under-promising and over-delivering is exactly what his pub tries to do.

We are not a rational animal but a rationalising one. Rory builds on Jonathan Haidt's image of the mind as a rider on an elephant: the elephant, wise, cautious and habitual, makes most decisions and cannot talk; the rider can do maths, talks a lot, and deludedly thinks he is in control. Rory's version: the conscious brain thinks it is the Oval Office, when in reality it is the press office, hastily cobbling together a plausible explanation for a decision taken somewhere else for reasons it does not fully understand. So consumers are not wilfully lying when they explain themselves, but research that only talks to the rider learns the press release, not the decision.

“The System 2 brain thinks it's the Oval Office, whereas in reality it's the press office.”

The point: Stated reasons are post-rationalisations, so work out what people feel rather than trusting what they say.

Variants: Examples vary: a firm raising prices blames raw materials because it cannot say greed; people wanted the euro first and then built the arguments for it.

Rory loves that American weather forecasts give two temperatures: the thermometer reading and the 'feels like' figure, which accounts for wind and humidity. As a self-described fat man of Celtic ancestry, he will not go to work in London at 85 degrees, yet happily strolls round Phoenix at 105, because it is a dry heat. He is not going outside to do chemistry; he needs to know what to wear. The body feels heat in proportion to the danger to health: humidity stops sweat cooling you, a breeze helps. Yet business, like the thermometer, measures whatever is numerically available and ignores how things feel.

“You should optimise for perception rather than objective reality because it's perception that determines behaviour.”

The point: Measures built around human experience predict behaviour better than objective physical ones.

Variants: He argues climate campaigners should have used a biological measure, the wet-bulb temperature at which even healthy humans cannot survive, and notes that among SI units almost only the lumen factors in human perception.

Farmers' markets defy economic logic: you buy direct from the producer, cutting out the middleman, yet pay more than at Tesco. To build one, Rory says, take a Tesco Metro and make it really inefficient. Put it outdoors in the rain, make people pay in seven places for seven things and queue seven times, like the Soviet Union. And people love them. Nobody would ever state that preference; the trend meets the spoken need for lower prices and more choice, which is Tesco Extra. But every trend breeds a counter-trend, and the farmers' market meets unspoken needs: street theatre, authenticity, relational exchange, and being seen to exercise discernment.

“To construct a farmers market is you take a Tesco Metro and you make it really inefficient.”

The point: When everyone optimises for efficiency, inefficiency itself can become the product, because counter-trends serve needs people never state.

Variants: A futurologist told him there are no trends, only vectors. In one telling the same logic explains airports: they all became shopping malls, 'Bluewater with some planes', so London City, with almost nothing there, feels incredible. Another adds a heuristic: a stall selling only dips must be good or it would not last six months.

A Michelin-starred restaurant in France wanted diners off their mobile phones. Most would post a bossy notice, and nobody paying a hundred euros for dinner wants to be bossed about; people hate it so much they invent hospitalised relatives to justify an exception. Instead, at the entrance, visible to everyone arriving but ostensibly addressed to those leaving, a sign read: 'Please remember to turn your mobile phone back on when leaving the restaurant.' It looks helpful rather than bossy, and implies an established social norm: sophisticated diners here would never dream of making calls during the meal. Rory says he would not have thought of it in 20 years on a desert island.

“Please remember to turn your mobile phone back on when leaving the restaurant.”

The point: An implied social norm, disguised as helpful information for others, gets compliance without the reactance a direct order provokes.

Variants: The location shifts (near Saint-Malo, possibly Saint-Émilion). In one telling it is Paul Craven's favourite ad, and Rory's moral is that for every logical solution there is a better one when you think backwards.

Drawing on Andy Clark's The Experience Machine, Rory explains that the optic nerve lacks the bandwidth for raw real-time imagery, so most of what we see is internally generated prediction, with the eyes used to update what is new, unexpected or moving. A JPEG or MPEG works the same way: rather than describing every pixel, as RAW does, each pixel has an expected value based on its neighbour, and data is spent only where reality differs. Without that compression your satellite dish would need to be the size of Jodrell Bank. The brain evolved the same data architecture as your Samsung television. So the expected is invisible, and the unexpected carries enormous weight.

“The human brain evolved exactly the same data architecture as Samsung televisions did, albeit over a longer time period.”

The point: Perception is prediction corrected by surprise, so small unexpected touches outweigh large expected ones.

Variants: One telling ends that advertising is the business of creating predictions and anticipation.

Rory plays a clip of the McGurk effect. The sound is always 'ba', but the face on screen mouths 'fa', and you hear 'fa'; close your eyes and you hear the real sound. A black box in the head decides it is more likely you misheard than misseen, and simply overwrites the sound. Borrowing George Loewenstein's phrase, he says that just as we have a sex drive and a food drive, we have a sense-making drive: evolution needs a decisive, unambiguous picture to act on, not an accurate account of each sense. This is where engineering thinking goes wrong: it assumes that to improve an experience you must improve the reality.

“Just as we have a sex drive and a food drive, we have a sense-making drive.”

The point: The senses are not independent channels, so experience can be changed through context without touching the objective thing.

Variants: One telling pairs it with a checkerboard illusion and attacks 'measurebators' who measure only the physical world.

Video calling is practically free and saves a fortune, yet for years it flopped. Rory's hunch is a framing error: it was sold as the poor man's air travel, not the rich man's phone call; the crap version of British Airways rather than the posh version of BT. That made it an inferior good, like margarine to butter or the pager to the mobile: what you do when you are not trusted on a plane. 'We won't send Sutherland to Frankfurt, he'll only raid the minibar, but he can have a pixelated conversation with Jürgen in the basement.' Had it been expensive and installable only in the chief executive's office, people would have been gagging to use it.

“It sold itself as the poor man's air travel, not the rich man's phone call.”

The point: Whether a substitute is framed as a downgrade or an upgrade decides whether it is adopted or stigmatised.

Variants: One telling adds practical failings (it should ring you, and the code should be a word, not 19 digits) and suggests branding it with a charity, so it reads as a choice rather than a compromise. He notes Post-it notes were marketed the expensive, exclusive way.

Riffing with Nassim Taleb on things we value for reasons other than the ones we give, Rory argues that the principal value of a dishwasher is not that it washes your plates. It gives you somewhere to put dirty plates out of sight, sealed so you cannot smell them. The biggest annoyance when it breaks down is not having to wash up; it is having to look at the plates. Likewise, the real value of a swimming pool is not swimming. It gives you licence to walk around your garden in a bathing costume on a hot day without feeling like an idiot. Taleb took the advice and bought a house with a pool. Nobody comparing dishwashers ever factors any of this in.

“The real value of a swimming pool isn't that you swim in it. It's that it gives you a licence to walk around your garden in a bathing costume.”

The point: Products often deliver their main value through an unadvertised side benefit, so ask what a thing is really for.

Variants: Rory uses it to ask what business you are really in, and to argue that much of what capitalism does is beneficial in ways no metric captures.

One of the first problems brought to Ogilvy's behavioural science practice came from a big continental European biscuit maker. It had cut the fat in its biscuits by about 20 per cent; blind tastings showed nobody noticed any difference, yet sales had fallen by about a quarter. Without pausing, the team asked: you didn't put 'now with lower fat' on the pack, did you? Of course they had. The sense-making brain looks for trade-offs: told something is healthier, it assumes it tastes worse, so it does. And some treats need an element of guilt to be enjoyable. To improve the nation's health, they should have cut the fat and told nobody.

“If you put 'new lower fat' on something, the brain will assume it tastes worse.”

The point: Announcing a virtue can trigger an assumed trade-off, so some improvements are best made by stealth.

Variants: In one telling the biscuit is Belgian and frames a wider point: 75 per cent of consumers say they will pay more for green products, but really about 8 per cent do, and many assume a green detergent will not clean well.

A remarkable number of scathing one-star TripAdvisor reviews begin: 'We had booked a table for my mother-in-law's birthday.' Rory reconstructs the scene. The man never wanted to go out; he would rather be in his pants watching the Champions League. He has been taken somewhere fancier than he would choose, he knows he will end up paying, and then he struggles to park. Then comes the litany: unfriendly service, slow food, they didn't wipe down the outside chairs. Nothing the restaurant did caused his mood, but he hunts for evidence to justify it. In other circumstances, the same restaurant would have given him a great time.

“We had booked a table for my mother-in-law's birthday.”

The point: People sincerely invent rational reasons for feelings that came from elsewhere, so stated complaints are a poor guide to their cause.

Rory closes his talks, and describes his own job, with one idea. Engineering fixes to physical reality are expensive and slow, and quite often the best way to solve a problem is to change the way people think about it instead. Rather than trying to change the world, change how people see the world. When people see the world differently, they behave differently, and that in turn changes the world. He uses it to tie together examples such as the Overground on the Tube map, the Uber map, the feels-like temperature and range anxiety, where reframing perception achieved what physical investment could not.

“Rather than trying to change the world, change how people see the world.”

The point: Shifting perception is often the cheapest route to changing behaviour, and so to changing reality.

Rory tells cafés to leave their tables and chairs on the pavement even when it is raining. People can see them from 300 or 400 yards and know there is a café and that it is open, because a closed shop would lock its furniture away from drunks. Fold them away and people assume you are closed. The elephant does not look for a sign saying open. His local coffee shop went bust under three owners on the same road, then thrived largely because the new owner put tables and chairs outside. A reader of Alchemy used it in reverse to deter late customers: just stack two chairs upside down on a table.

“The elephant doesn't look for a sign that says open.”

The point: Small environmental cues signal availability more powerfully than explicit messages, and absent cues are read as rejection.

Variants: In one telling a shop open till 10pm hid its lights behind price posters, so nobody could tell it was open. Rory notes that tea rooms with odd hours suffer most, since customers take any clue that you are closed.

Every country has a contextual drink. Sipped in France, Pernod seems the most sophisticated drink imaginable, so you buy a large bottle to take home. The moment you pour a glass in Britain you wonder why on earth you bought it. Hungary has Unicum; the Greeks, Rory jokes, have managed Retsina, which tastes awful even in Greece. Rosé tastes better by the sea and Guinness better in Ireland, though chemically identical. This is not a freak of psychophysics: setting is part of the product. Maybe Greek restaurants with a huge picture of the Acropolis know something: Greek food tastes better if you recreate the atmosphere of Greece.

“Pernod tastes delicious in France but tastes like piss if you bring a bottle home.”

The point: Much of a product's enjoyment comes from its setting and complements, so recreating the right context can improve the product.

Variants: One telling pairs it with popcorn, which is awful without a film; another uses it to introduce contextual, location-specific communication as a nudge.

A drug has a chemical effect and a placebo effect, and both aid healing. Yet drug trials are designed to subtract the placebo from the total so the chemistry can be proven in isolation, and Western pharmacology wants to distil out the chemistry and discard the psychology. Rory thinks that an enormous mistake. If both work, the most effective treatment is drug plus placebo, so why not see it as placebo times medicine equals overall effect, and research how to maximise the placebo as hard as the pharmacology? Science is uneasy because it cannot fully explain the placebo, but nobody knew how aspirin worked until recently either.

“Medicine spends all its time trying to subtract a healing effect from another healing effect.”

The point: Psychological value should be amplified alongside intrinsic value, not discounted as contamination.

Variants: One telling starts from a researcher's beta blocker that worked on 80 per cent of people against 40 per cent for the placebo, notes that Japan welcomes the placebo as part of efficacy, and adds that banning everything found by luck would have cost us penicillin. Another predicts weight loss will be solved by a drug plus a behavioural regimen.

Rory's accusation: deep down, everybody over about 35 wants to live in suburbia, with a bit of lawn and somewhere to park, but in Britain suburbia isn't cool. Move to Bexleyheath and your friends disown you; he could not get his younger colleagues to move to Bromley with a gun to their heads. Its image is terrible, but the actual utility of living there is fantastic: completely uncool, but utterly brilliant. That, he thinks, is why Brits love Los Angeles. Buy a glorious suburban house off Mulholland Drive and, because it is LA, it is totally cool.

“It's completely uncool, but it's utterly brilliant.”

The point: The same choice carries different status in different contexts, and changing the context can unlock a suppressed preference.

Variants: He adds that for Brits the most pointless American cities to visit are New York and San Francisco, poor attempts at a European city, while Austin, Houston, LA and Chicago offer a different life.

To the British, if the sun is shining it is a nice day, whatever the temperature. Rory arrived in Johannesburg in their winter to a glorious, cloudless blue sky, lovelier than any British summer's day. The Ogilvy people collecting him, blowing on their hands, apologised that he had to come during such terrible weather. He burst out laughing, thinking they were joking. They were not. In a Johannesburg winter there is never a cloud for six months, so blue sky is standard issue; what they meant was that it was unusually cold, with frost overnight and two or three degrees. Good and bad weather, like price, are judged entirely by local comparison.

“I'm sorry you had to arrive during some weather.”

The point: Judgements of good and bad are relative to local expectation, so the same conditions can feel good or bad depending on the frame.

Variants: One telling adds Portuguese locals in puffer jackets while British tourists stroll in T-shirts.

Rory, or in one telling a friend, went on YouTube to find out how to thaw a frozen boiler condensate pipe. The eight-minute video is, 99.99 per cent of the time, the most boring footage in the history of the world, almost a Monty Python idea of bad television. You would never watch it for entertainment. But at the moment your boiler exit pipe has frozen over, it is better than Citizen Kane, and no detail is irrelevant. He credits YouTube with becoming the instruction manual for millions of devices, replacing the 40-page booklet you always lost.

“When your boiler exit vent freezes over, it's better than Citizen Kane.”

The point: Value depends on the viewer's situation, so content worthless to most can be priceless to the right person at the right moment.

Variants: The film it beats changes: Citizen Kane, The Godfather Part II, or Last Year at Marienbad.

Ask tennis fans what makes it so good and they will say it is fast, exciting, televisual. Rory bets not one in 500 would say the scoring system, yet that is the work of genius. Because points are grouped into games and sets, with serve held and broken, the losing player always has a route back and the match has a narrative arc, while an uneven match ends quickly. Score it like basketball and it would be unwatchable: 'Federer leads Nadal 106 points to 84', and everyone goes to the pub. Rory agrees with the basketball critic who suggested both teams start on 100 points and play the last five minutes.

“If you played tennis but you scored it like basketball, it would be unwatchable.”

The point: The structure framing an activity, often invisible to the people enjoying it, can create most of the experience.

Variants: One telling credits the insight to his friend Richard Thomason and adds that baseball tames its randomness by settling the World Series over seven games.

Ford's Kumar Galhotra says making a car is a hundred thousand rational decisions in search of one emotional decision. Volkswagen spent enormous sums yet for years sold amazingly few cars in the United States, about 0.5 per cent of the market by Rory's account, because German engineers refused to fit cup holders. Anyone who knows American car buying knows the number of cup holders comes second or third in the decision tree, well above cornering ability. The refusal was cultural: drive-throughs barely work in Germany, because most Germans will not allow food or drink in the car. Your car is a temple, and defiling it with a bargain bucket is unthinkable.

“If you don't win the emotional decision at the bottom, you don't sell the car.”

The point: A trivial-seeming detail that matters to the buyer's culture can outweigh a hundred rational strengths.

At Wagamama, staff ask first-timers whether they have been before, then explain that it is based on an authentic Japanese noodle bar, so food comes fresh from the kitchen as soon as it is ready and not necessarily in the order you expect. Rory thinks that one sentence is pivotal to the chain's survival. Random order massively helps kitchen efficiency, but without the explanation half the diners would see a gyoza arrive after a vast bowl of soup and decide the staff were useless; in a French restaurant, soup at the end would be intolerable. Japanese people have told him it never happens in Japan. Claim to be Japanese and people will accept almost any weirdness.

“If you change the frame you can change judgement.”

The point: Explaining a quirk in advance, with a flattering reason, turns an operational convenience into a perceived feature.

Variants: Rory jokes that Argos's greatest tragedy is not being Japanese: a few bonsai and a pebble would make it the coolest shop in the world. He suspects ceremonial-grade matcha could have been invented by a marketer.

For a time Coca-Cola, possibly to save money, put red caps on Diet Coke, which had always had white or silver ones. Rory rang the agency handling Coke to tell them they were insane. After years of reaching for the silver lid, a regular Diet Coke drinker finds the red cap makes the drink almost undrinkable, and people would also assume it was the high-calorie version. He predicted they would revert within months, and they did; he suspects sales had dropped 10 or 15 per cent. We conflate taste with sight and smell because it is more efficient for the brain.

“Diet Coke with a red cap tastes disgusting. It's as extreme as that.”

The point: Visual cues are part of taste, so changing the packaging can change the product people experience.

Among direct marketing experiments, Rory cites one in which putting a photo of a smiling person on the phone next to the phone number had a larger effect on loan applications than reducing the APR. He uses the idea to sum up behavioural economics: the context, medium and interface in which a choice is made can outweigh its long-term consequences, even when people know those consequences. A lender could win more sign-ups by making the application attractive than by cutting the interest rate. People unaware of such findings spend careers being logical and convincing to colleagues, and the very logic that makes them look a safe pair of hands damages the business.

“Putting a smiling face with someone on the phone next to the phone number would have a larger effect on loan applications than reducing the APR.”

The point: The design of the moment of choice can outweigh the real costs and benefits of the options.

Variants: In one telling it is a 'villainous' illustration about an attractive credit card application form rather than a tested photo.

When Cadbury changed the shape of the individual chunks in its bars, making them rounder and more fun, it was bombarded with complaints that it had changed the recipe and the chocolate had got sweeter. The formulation had not changed at all. Rory uses it as evidence that the brain does not taste in isolation: rounder food tastes sweeter, so change the shape of something and you change its taste. That may explain why children's lollipops are round, since children want something bonkers sweet. He adds, as an aside, that the Chupa Chups logo was designed by Salvador Dalí.

“If you change the shape of something, you change the taste.”

The point: Form and presentation alter sensory experience itself, not just appeal, so a cosmetic change can read as a product change.

A bank asked Rory whether an advertising campaign could encourage online banking. He said it was easy: the last thing people want to see when they log in is their balance. He has friends who avoid their own bank's cash machines in case they show it. Why willingly expose yourself to bad news? Make 'Tell me my balance' an option rather than the default, and he reckons twice as many people would log on, three times as often. It would cost about 50 quid, yet it never happens. He calls it one of his most banal pieces of advice to banks.

“The last thing in the world you ever want to see is your balance.”

The point: Removing a small, unwanted source of anxiety from an interface can change behaviour more than an advertising campaign.

Variants: In one telling it follows Morrisons' failed 'Match and More' scheme, which told shoppers they would have saved money elsewhere: people may simply prefer not to know.

Concentrated detergents are kinder to the environment and cheaper to transport, but sell a plain concentrate and people either overdose or think it's no good, because of the heuristic that more is better; using less feels like doing less of a job. We also believe that the more intricate something is, the better it works. So cut the volume and make up for it with complexity. Coloured flecks make a powder seem more effective because it has two things in it. Finish Quantum dishwasher tablets are insanely intricate: coloured powders arranged around a 'power ball'. Nobody notices they are using less, and System 1 says: that's really complicated, so it must be great.

“That's really complicated, so it must be great.”

The point: One heuristic (intricacy means quality) can be used to override another (more is better) and deliver a hidden benefit.

Google Maps' public transport mode assumes that if you use public transport, it is because you have no car. Ask for a route from Kent into central London and it sends you by bus to the station and on foot at the other end, when what everybody actually does is drive to the station and catch a train. That very British practice, also common in Connecticut, is invisible to it: in Los Angeles you only take public transport if you have no car, and Google is very Californian in that assumption. You cannot search for a mixed car-and-train journey, so it repeatedly gives ludicrously pessimistic journey times for trains.

“Google is very Californian in that assumption.”

The point: Software encodes its designers' cultural assumptions and quietly misleads users who live differently.

When Google started, every tech company thought you had to build a portal with weather, sports scores, news and a bit of search. Google stayed a single bar with 'I'm Feeling Lucky'. It was a technological success but also a psychological one: instinctively, as with the jack-of-all-trades heuristic, we assume something that does one thing is better at it than something that does many. Ayelet Fishbach calls this goal dilution. Everyone knows it from TV shops, where the combined TV and DVD players sit despised at the shabby end. We know nothing of their quality, but assume a crap telly and a rubbish DVD player, so we buy one of each.

“If you're just a search engine, people assume you're a very, very good search engine.”

The point: Doing one thing signals expertise, while adding functions can make each look worse.

Indian restaurants are usually dead at lunchtime, but Dishoom is busy. Rory thinks it cracked the problem by serving breakfast. Because people are already inside at 11, 11.30 and noon, passers-by instinctively feel it is fine to go in. People dread being the first customer and getting trapped by social obligation; in a café you can order just a drink, while in an empty restaurant you cannot. Dishoom also puts tables and chairs on the pavement, which say from 500 yards away that there is somewhere to eat, that it is open and that it suits lunch. He once advised an empty jewellery shop that three of its four staff should pretend to be customers.

“People are terrified of being the first customer in a store.”

The point: Visible occupancy and cues of openness, not menu or price, decide whether people walk in.

Rory imagines a world where all cars are electric and a rogue Volkswagen engineer proposes something new: put a tank of inflammable liquid in the car, engineer explosions in cylinders, add a gearbox, oil, filters and some 250 moving parts. Is it quieter? No. Cleaner? No, it farts out the back. Can you refill at home? Obviously not; you go to a special place that sells Ginsters. Only the tiny fraction who drive 350 miles a day without stopping would benefit. 'I don't really see this is going to fly.' A hundred years of brilliant refinement have made the combustion engine familiar, but to a physicist it is inherently ridiculous.

“If all cars were electric and someone invented the internal combustion engine, people would look at them as if they were completely mad.”

The point: Inverting the order of history exposes how much of our preference for the incumbent is familiarity, not merit.

Variants: One telling, from an Evening Standard piece, adds that Top Gear drove electric cars until they died, which you could do with any petrol car.

In the Netherlands, someone took KFC or McDonald's chicken, put it on white china plates, dressed it with fresh vegetables or a fancy salad, and served it at an organic food fair in Amsterdam. Visitors praised its wholesomeness, originality and authenticity. Rory says the Austrian economists would ask what is wrong with that. Creating value through story and presentation is efficient and even environmentally friendly: you don't have to make anything, chop anything down or burn anything, just tell a story around something that already exists. Assuming we judge things independently of how they are displayed is, he says, the worst misconception in problem solving.

“The idea that our evaluation of something is independent of the context in which it's displayed is the worst misconception in human problem solving.”

The point: Presentation and setting shape quality judgements as much as the product, so changing context is a cheap, low-impact way to create value.

Rory makes a mischievous claim certain to scandalise some listeners: miso soup isn't really very nice. If it didn't exist and your child said they had invented a new soup, with a leaf and a white cube floating in it for no apparent reason, you would have a sip and say don't give up the day job. Likewise, hear of an Amazonian tribe reduced to eating bits of raw fish wrapped in rice and you would think: poor things. He drinks a lot of miso himself; it is great because it is Japanese and carries those associations. That is not a con: provenance genuinely adds to enjoyment.

“To what extent are we actually eating what we eat and to what extent are we consuming stories?”

The point: Provenance and story are part of the taste of food, not a deception layered on top.

If you came in from potting plants with muddy hands, you would not rub them with dry paper; you would use water. Most of the world, including the Islamic world and South-East Asia, finds wiping with dry paper repellent, yet the West sticks with it. Working for Kimberly-Clark, Rory's team noticed the shelf: half an acre of dry paper and two small packs of moist on the top shelf. The brain infers that dry is mainstream and moist is for weirdos and people with a medical condition, and defaults to dry. The fix is what Walkers did with crisps: launch many variants to dominate the shelf.

“That thing for weirdos and people with a medical condition up on the top shelf.”

The point: Shelf space acts as social proof, signalling which option is normal, so share of shelf shapes choice.

Variants: One telling notes that Deadpool 2's writer put an elaborate comic rant about dry paper into the film.

Australians long suffered from the cultural cringe, convinced that the English patronised them as colonials with corks hanging off their hats. That was true in the 1950s and 60s. What they failed to realise, Rory says, is that the British had since spent a decade or more watching Neighbours and Home and Away. By then nobody was patronising anybody: the mainstream British reaction to meeting an Australian was envy and jealousy, you lucky so-and-sos. He suspects a residual belief persists in Australia that Britons still look down on them.

“Nobody was patronising anybody by that stage.”

The point: Popular culture can shift how one group sees another far more than elite opinion, and the sense of being judged lags behind the change.

A brilliant Ogilvy copywriter pointed out to Rory that in New York people speak fast and in the American South they speak slowly, and both are forms of politeness. In New York, speaking quickly acknowledges that the other person's time is valuable. In the South, speaking slowly shows how much of your own time you are willing to give them, intimating that they are worth it. The same message is sent in completely opposite ways. Rory treats it as one of many perceptual paradoxes, and links it to Cialdini's observation that things sell by opposites: because everyone has one, or because hardly anyone does.

“In New York people speak fast, in the American South they speak slowly; both of them are a form of politeness.”

The point: Opposite behaviours can carry the same meaning in different contexts, so there is no single right way to signal a value.

Nobody revolts against what they see as inevitable. In 1988 the landline, the fax and the computer were all in the office, so nobody grudged commuting. Nobody minded buying whole CDs until downloads existed and record companies still insisted on them. Once people had experienced putting a cardigan over their pyjamas and clicking 'join meeting', the same train journey felt ten times more painful, especially if you go in only to do emails and Zoom calls. Rory struggled to persuade his wife to get a Quooker boiling-water tap; now waiting for a kettle feels weirdly Victorian. Once you have experienced the new behaviour, the old one seems ridiculous.

“Nobody revolts against anything they see is actually inevitable.”

The point: The existence of an alternative transforms how a burden feels, so the hard part of adoption is the first trial, after which reversion is rare.

Variants: One telling adds that electric car buyers rarely go back; the other suggests flexible working could be hourly, starting at 10.30 on an empty off-peak train.

We think Olympic swimmers are really fast because they are fast compared with other swimmers. As a mode of human transport, swimming is hopeless: Rory calculated a world-record swim at around 4 mph, slower than someone in a pair of wellies. That is why broadcasters make sure nobody walks alongside the pool, since even world champions could be outpaced by a pensioner with a Zimmer frame putting on a burst. He reckons he could outrun them in wellingtons while hopping. Whenever we judge anything we reach for a frame of comparison, and the frame we are given can produce utterly nonsensical conclusions.

“I could outrun major Olympic swimmers in wellingtons while hopping.”

The point: Judgements depend on the comparison set, and changing the comparison can transform how something is perceived.

If you doubt that placebo choice exists, Rory says, consider lifts. Many 'door close' buttons are not connected to anything: they do not speed up the doors at all, but they give irritable, impatient people something to jab so they feel in control. Asked whether this is true or a rumour, he confirms it and calls it the illusion of agency. He puts it alongside approximate car positions on early ride-hailing maps and stylised delivery trackers. Restaurants do the verbal version: 'still or sparkling?', 'red or white?', even 'tea or coffee?' give the illusion of control while quietly freezing off options such as tap water.

“It's placebo choice: it gives you the illusion that you're in control.”

The point: A sense of control reduces frustration even when the control is illusory, and a narrow either-or question can feel like freedom.

Royal Mail delivered about 97 or 98 per cent of first-class post the next day and nearly broke itself trying to reach 99. Who would notice? Worse, asked to guess, the public said somewhere between 40 and 70 per cent. It was great food in a restaurant that smelled of sewage. If your perception is much worse than your reality, Rory asks, why spend a fortune changing the reality? Fix the perception first. A rational, System 2 campaign would quote the 97 per cent. The System 1 campaign would say more first-class mail arrives next day in Britain than in Germany, which to a British mind means good enough. To please Britons, tell them they do it better than the Germans.

“If your perception is much worse than your reality, what on earth are you doing trying to change the reality?”

The point: When perception lags reality, closing the perception gap is cheaper and adds more value than improving the product.

A US retailer, Rory thinks Best Buy, made shoppers sign in or register before they could buy. It added a button: 'Check out as guest. Don't worry, you'll have the opportunity to register later.' Sales rose by hundreds of millions of dollars, and the vast majority of guests registered at the end anyway. Only the order had changed; the effort was identical. Typing your name and address framed as 'registration', before you have bought anything, feels like a maddening imposition. Typing it so your fridge can be delivered feels reasonable, and adding a password to details you have already given feels like nothing.

“It's objectively similar but subjectively different, and that's where to look for 10x's.”

The point: The sequence and framing of an identical task can transform how burdensome it feels.

Variants: The figures vary by telling: $200 million with 87 per cent registering afterwards, or $300 million with 97 per cent.

Rory shows Frans de Waal's fairness experiment. Two capuchin monkeys happily trade rocks for cucumber, again and again, until one starts receiving grapes for the same task. The cucumber monkey tests her rock against the wall, takes the cucumber and throws it back at the trainer. De Waal calls it the Wall Street protest. Rory's reading: the monkey thought it a perfectly fair deal and liked the trainer. Change only the context, with a neighbour getting grapes, and the trainer goes from friend to enemy. The same effect can make people miserable or happy, and successful businesses understand we are not utility-maximising monkeys, otherwise we would be happy with the cucumber.

“We're not utility-maximising monkeys, otherwise we'd be happy with the cucumber.”

The point: Satisfaction depends on relative treatment, not absolute reward, so changing the comparison changes how the same deal feels.

For a charity appeal, Rory's Ogilvy colleagues sent 100,000 envelopes in each of four variants: one saying volunteers had delivered it, one prompting a form to claim a 25 per cent tax rebate, one on better-quality paper, and one in portrait format. Logically only the tax rebate should matter. It cut donations by over 30 per cent and even reduced the share of people giving at all. The other three each raised donations by over 10 per cent, and the better paper attracted more large gifts. Perhaps a flimsy envelope feels wrong for a big cheque, or people reciprocate: they sent me nice paper, I owe them fifty quid. Only testing could have found it.

“The human mind does not run on logic any more than a horse runs on petrol.”

The point: Behaviour responds to cues economists dismiss as irrelevant, and the rational lever can backfire, so test rather than theorise.

Variants: One telling is about the paper alone, for Christian Aid Week, with more donations over £50; the other is a hurricane appeal with more gifts of £100 or more.

An old barman's trick that works on heavy drinkers in golf clubs: after a couple of real gin and tonics, serve plain tonic with ice and lemon in a glass whose rim has been dipped in gin, so they get the smell. Regular drinkers often fail to notice and carry on acting drunk, though their blood alcohol is zero; the brain takes a shortcut: every time I do this I start talking bollocks and swaying around. Rory tells of an unnamed high court judge breathalysed leaving a golf club. The test came back clear, saving his career. He immediately went back and demanded the barman be fired for watering down the drinks, somewhat ungratefully.

“He immediately went back and demanded the barman was fired for watering down the drinks. Somewhat ungratefully, I would argue.”

The point: Expectation and sensory cues alone can produce the behaviour we attribute to a substance.

Rory saw an analysis of customer feedback from a hotel in Memphis that asked guests about many things, including an irrelevant ice sculpture in the garden in winter. Virtually every score depended on the check-in experience. Guests with a bad check-in constructed a narrative of a bad hotel and went looking for corroborating evidence; those with a good one decided to be in a good mood. Those who checked in quickly thought the ice sculpture was brilliant; those who had waited 20 minutes behind a United Airlines flight crew rated the sculpture and the garden as poor, or just okay.

“They thought the ice sculpture and the garden was crap if they'd had to wait 20 minutes at check-in.”

The point: First impressions colour judgements of unrelated things, so the opening moment of a service matters disproportionately.

When St Pancras reopened, Freud Communications mentioned in every press release that it had the longest champagne bar in Europe. Rory finds the superlative absurd: nobody chooses a champagne bar for its length or complains one was too short, and it isn't even that long; he expected to see the curvature of the earth. Yet that one sentence makes you want to go and look, and reframes the station from a utilitarian transit hub into a destination and meeting place. He calls it benign, brilliantly irrelevant nonsense. A billion pounds was spent rebuilding London Bridge station; he would have added the largest florist in Europe so that people would love it.

“It's a brilliantly irrelevant wacko thing.”

The point: A memorable, even silly, claim can give a functional place an identity that multiplies appreciation of what was built.

Driving back from Deal, Rory kept missing the turn from the M20 onto the M25: four or five times out of a dozen. Once is your fault; four times, there is something wrong with the junction. On Google Street View he found why. The first gantry gives no primary destination such as Brighton or Gatwick, only Swanley, which nobody can place, so you wait for the next sign on the horizon. By the time you can read it, the lanes have divided and the exit has gone; some drivers swerve across the hatching. He told the Highways Agency they might as well paint 'Haha, you've missed the turning'. They met him, agreed, and changed it.

“By the time you can read that sign, it's too late to act on the information it contains.”

The point: When a system asks for a choice before giving the information to make it, people blame themselves rather than the design.

Behavioural scientists fret about the replication crisis. Rory says it is not a bug but a feature. Citing Bent Flyvbjerg, he argues that social science's quest for universal, context-free laws like gravity may undermine its usefulness: snooker balls move the same way everywhere, people do not. Findings fail to replicate precisely because context has such a huge effect on how people think and act; the same person facing the same choice in a different context decides completely differently. For marketers, those tiny contextual differences are the interesting bit. It is also why a business cannot import a finding from elsewhere and must test it on its own customers.

“If anybody did discover a universal law of human behaviour, the marketing community would have to have them killed.”

The point: Because behaviour is context-dependent, findings must be tested locally, and finding the right context is the marketer's real job.

Rory sets out his model of decision-making as a chain. There is the objective thing; then the context in which it is perceived; the perception in that context; the meaning we derive from the perception; the emotion generated by that meaning; and finally the behaviour the emotion drives, followed by an act of furious post-rationalisation. In physics one thing maps neatly onto another. In human behaviour, with four or five messy intermediate stages, the mapping is anything but neat. Change the story about something and you change its meaning, our visceral response and our evaluation: it becomes a different thing. So objective reality matters far less than people assume.

“Things are not things. We don't actually respond to things, we respond to things in a context from which we derive meaning.”

The point: You can change behaviour by working on context, perception or meaning without changing the object itself.

Rory shows an optical illusion with two squares, A and B. Everyone agrees A is dark and B is light, because the brain assumes a shadow falls on one and compensates by making it look brighter. Cover the join, or change the context, and you see they are exactly the same grey. Yet once the context returns, it is impossible to see them as the same again. Almost all our senses, and our sense of pleasure, pain and value, operate relatively rather than absolutely. Economics assumes we value things unaffected by what sits alongside them, which Rory calls dangerously wrong: reality does not map neatly onto emotion, so stop measuring reality alone.

“Our perception of virtually anything operates at a relative level, not at an absolute level.”

The point: Perception and valuation are relative to context, so measuring objective reality alone misses how people experience it.

Rory calls the electronic cigarette a ten-times innovation. After years of fiddling with nicotine patches, the discovery was that smokers are not really addicted to nicotine: more and more vapers drift to the zero-nicotine variant without really noticing. What the brain was attached to was the act: inhaling something quite hot that hits, he thinks, the trigeminal nerve at the back of the throat, then breathing out and relaxing. His trick for finding ten-times innovations follows from it: look for things that are objectively similar but subjectively different, or vice versa.

“Look for things which are objectively similar but subjectively different, or vice versa.”

The point: Habits often attach to the sensory ritual rather than the active ingredient, so the ingredient can sometimes be removed.

If there is one thing to know from behavioural science, Rory says, it is this: we believe we pay attention to what is important, when in fact we think important whatever we are paying attention to. You cannot control what you feel is important, but you can control where you direct your attention, and so hack your own priorities. The same lever works on whole societies. As Citizen Kane has it: make the headline big enough and I'll make the story big enough. Political bias in news is less about what to think than what to think about. American truck makers used the same emissions cheats as Volkswagen, but trucks never left the trade press.

“The way our brains really work is we think important what we're paying attention to.”

The point: Control of attention is control of perceived importance, so whoever sets the focus sets the priorities.

Rory asks, as a naive ad man, what is actually wrong with placebos, which are effectively banned as treatments in the UK. They seem fantastic to him. They cost almost nothing to develop. They demonstrably work, sometimes extraordinarily well. And they have no side effects, or if a patient does report side effects, those must be imaginary too, so they can safely be ignored. The joke sets up his wider argument that we undervalue and distrust interventions that work through the mind.

“They have no side effects, or if they do, they're imaginary, so you can safely ignore them.”

The point: Interventions that work through perception are cheap and effective, yet institutions treat them as illegitimate.

Every culture has an unconscious norm for two people approaching each other on a pavement: which way to step and how early to swerve. In the UK both veer left at about three feet; in India it is more like two and a half feet, because people are happier being close; in Korea they go right. Airports gather people from every culture, each following a different rule, so nobody knows what anyone else will do and crowd movement becomes fundamentally dysfunctional. Rory also suspects people never quite recovered the knack of walking in crowds after Covid.

“Everybody's operating unconsciously to a completely different rule.”

The point: Invisible coordination norms only work when everyone shares them, and mixing them causes chaos.

Variants: One telling adds a famous mid-air collision in which, beyond an air traffic control error, two conflicting rules about which way to evade were the real cause.

J. P. Morgan said every man has two reasons for doing something: a good reason and the real reason. In most corporate settings, asking why people clean their teeth would get you treated as a lunatic: the official answer is dental health and fewer cavities. But look at when people brush: always first thing in the morning and before a date, almost never after lunch. It is about vanity and fear of bad breath. Further proof: if it were about gums, why would about 95 per cent of toothpastes be mint-flavoured? The dumb question exposes the gap between the ostensibly logical reason and the real one.

“Every man has two reasons for doing something: a good reason and the real reason.”

The point: Observe behaviour rather than stated motives to find the real driver you can design for.

Variants: The mint figure is 95 per cent in one telling and 97 per cent, confirmed by someone at Colgate, in the other.

Rory admires the restaurant marketer Mark McCulloch, who says restaurants assume passers-by know far more about them than they do: the owners know what they sell, but people walking past do not. McCulloch found a woman who, every lunchtime for years, walked about 200 yards past a branch of the salad chain Chop'd on her way to buy a salad somewhere further away. Chop'd did not say salad on the outside. She saw the name, associated it with chopsticks, assumed spicy food, which she didn't like, and ignored it. Nobody had been explicit. Ogilvy staff realised many people would not even know that Wahaca was Mexican.

“There's an amazing amount of money you can make just by explaining yourself very clearly.”

The point: A business can quietly lose customers to a misreading it has never noticed, so state the obvious plainly.

Variants: Seven years in one telling, nine in the other.

Hotels proudly tell you they've had a refurbishment programme costing $400,000 per room. And you think: to make it look like every other hotel room I've ever stayed in. Following Andy Clark's predictive-processing point, it's the weird stuff we notice, the things we didn't expect, that really chime with us. Matching the standard is invisible.

“To make it look like every other hotel room I've ever stayed in.”

The point: Spending to meet expectations goes unnoticed; only the unexpected registers.

Put 'kind to the environment' on a detergent or washing powder and the System 1 part of the brain automatically infers that it is not very good at cleaning your clothes. The human brain is programmed to look for trade-offs: if something is better for X, it is probably worse for Y. That is why virtuous claims can backfire and good is sometimes best done by stealth.

“The human brain is programmed to look for trade-offs. If something's better for X, it's probably worse for Y.”

The point: Consumers assume every gain has a hidden cost, so a virtue claim can undermine the product's core promise.

We get under-excited by online audiences. When Nudgestock streamed live during Covid to 120,000 people, some said it could have been bigger. Rory replied that's one and a half Wembleys: present to a full Wembley Stadium and you'd call it the biggest event of your life. During Covid he did lots of podcasts, some tiny, reasoning that reaching 600 people in an hour beats reaching one. Then a film student called Hugo started clipping and 'TikTok-ising' them, and now he can't walk to St Pancras without a selfie request.

“Hold on, that's one and a half Wembleys.”

The point: Reframing a number against a vivid physical reference changes how impressive it feels.

After 36 years in advertising, Rory knows you can present the same information 15 different ways and get completely different emotional responses depending on what you leave out and highlight. Say 75 per cent of people will survive or 25 per cent will die: the same fact, but people respond differently because there is implicit social context in the framing. There is no point simply giving people objective information, because we are incapable of consuming it objectively. He praises New South Wales's government behavioural science unit.

“There's no point in simply providing people with objective information when they won't consume it in an objective fashion.”

The point: Framing is unavoidable, so communicators must choose it deliberately.

Rory notes that in the Netherlands people do not have curtains on their downstairs windows, and regard anyone who does as secretive and up to no good. So if you take an after-dark train journey through the Netherlands, it is like a moving version of Hitchcock's Rear Window.

“It's like a moving version of Hitchcock's Rear Window.”

The point: Norms about privacy are cultural: what signals honesty in one country signals secrecy in another.

Rory's grandfather was a GP in Tredegar, working for the miners' medical cooperative where Nye Bevan got the idea for the NHS. He admitted that before antibiotics a GP was about 30 percent doctor and 70 percent shaman: much of the value came from reassurance, jollying people along and giving them confidence. Rory thinks that is fine, and that Ed Miliband's much-mocked 'Ed Stone' got one line spot on: 'an NHS with the time to care'.

“Before antibiotics came along, as a GP you were about 30 percent doctor, 70 percent shaman.”

The point: Feeling looked after is a large part of what healthcare delivers.

Rory quotes a passage from Bent Flyvbjerg's book drawing on a French anthropologist, Bourdieu [?]: context is basically everything. Gift-giving is generally a nice thing in the right context. But if someone gives you a present and you immediately return it, that is not a gift, and it is not even neutral: it is an insult. Tiny contextual differences make a huge difference to how things are perceived.

“If someone gives you a present and you return it, that's not a gift. It's not even neutral, it's an insult.”

The point: The same act can flip meaning entirely with a small change of context.

Closing his welcome, Rory asked the audience how many wanted the event to become annual, framing the show of hands explicitly as a commitment device. When the hands went up he announced 'consistency bias, good, marvellous, got you to raise your hand, all done', a live demonstration that a small public act binds people to future behaviour.

“Consistency bias. Good. Marvellous. Got you to raise your hand. All done.”

The point: Getting people to make a small public commitment makes them more likely to act consistently with it later.

Rory flew to the event on an Embraer 190 from London City, a plane costing perhaps £55 million. The only thing he really remembered about the flight was that someone had spent probably £25 putting a mobile phone holder on the back of the seat in front. Marketers are unpopular in organisations because they have no sense of proportion, but neither do consumers: we can be hugely impressed by utterly trivial details and ignore things millions have been spent on. Yet at British Airways the person responsible for seatback fitments is fairly far down the food chain.

“Marketers have no sense of proportion. But the point is consumers have no sense of proportion either.”

The point: Customer delight often comes from cheap, trivial details rather than the expensive engineering organisations prize.

Rory finds it fascinating that the tea plant, Camellia sinensis, is the same everywhere. Setting aside things like rooibos, matcha, lapsang souchong, Earl Grey, green tea, black tea and white tea all come from the same plant. Everything that makes them different is how the leaf is treated.

“It's genuinely one goddamn plant.”

The point: Processing and framing, not raw material, create most of the variety and value in a category.

Search Amazon for a Samsung television and the first three results are similar sets from makers Rory has never heard of. No retailer would do that: it destroys the premium sale. His Samsung suddenly looks expensive, he won't risk 500 on an unknown brand, so he buys nothing or goes to John Lewis, where choice architecture makes sense. Shops put some products on a pedestal; an egalitarian grid has sovietised retail. He thinks tech and consulting are conniving with managerial bureaucrats to make capitalism increasingly Soviet.

“By putting everything in an egalitarian grid you've kind of sovietised retail.”

The point: Flat, undifferentiated presentation removes the cues people need to choose and pay a premium.

American and Asian tourists want a London hotel to be a bit London-y, with a man in a top hat at the door. A friend of Rory's booked some very cool friends from Los Angeles into the Hempel, an LA-style hotel in Bayswater. They were gutted: if they come to London, they said, they want horse brasses and hunting prints. Such nuances are easy to lose when only costs are counted.

“If I come to London I want horse brasses, I want hunting prints.”

The point: Customers often pay for local character and status cues that efficiency-minded design strips out.

Rory asks whether an ancient Greek football commentator would be more honest than a modern one. Rather than attributing every result to human intent, he would acknowledge fate: 'Everything was going well for United, and then Phoebus Apollo decided to thwart their midfield efforts.' Citing Chesterton, that when we stop believing in God we believe in anything, Rory argues that invoking capricious gods may be a more accurate picture of a complex, unpredictable world than seeing intentional agency in every outcome.

“Everything was going well for United, and then Phoebus Apollo decided to thwart their midfield efforts.”

The point: Acknowledging luck and the unknowable can be more truthful than narratives of total human control.

An Onion parody about upper-middle-class women in Marin County taking up candle making makes Rory's point. In the 18th century candle making was necessary domestic drudgery; the moment you could buy ready-made candles, it became an extraordinary pleasure. Bread making is the same. 'Artisan' basically means manufactured through an inefficient process, and it is all the rage.

“Artisan basically means manufactured through an inefficient process.”

The point: Once necessity disappears, drudgery can be reframed as pleasure and inefficiency sold as a premium.

Most pollsters got the 2024 US election wrong, but a WPP colleague had marketing data showing Trump's 'voltage' on the Brand Asset Valuator was far higher, and called it for Trump five days out. One of the cleverest methods asked people not how they would vote but how they thought their neighbours would. Asked about ourselves, we think about what the answer says about us. Offered canapés with caviar or KFC, you will say caviar, yet walking past the bargain bucket you turn left.

“Our question is what does it say about me to answer this question, not what's the honest answer to this question.”

The point: Indirect questions strip out self-presentation and reveal real preferences.

Rory calls himself an electrification enthusiast. Every British car traditionalist resisted automatic transmission until they drove one and never went back, just as everyone scoffs at air fryers until they own one. Everything in a petrol engine, cylinders, gearbox, spark plugs, filters, exists only to rotate a shaft, and almost every other rotating device has moved to electric motors. He cites the car industry mantra that the best part is no part, and is most excited by small electric car-like objects such as the two-seat Microlino bubble car.

“Every car traditionalist in Britain resisted automatic transmission until they drove one, at which point they never went back.”

The point: Resistance to a new technology often vanishes on first use, so trial matters more than argument.

Rory recently learned there is a whole cohort of people who avoid Marks and Spencer not because of its prices or products, but because they don't like the people who shop there. He confesses he is a bit like that with Gail's bakery: he quite likes the product, but not the people in the queue.

“I really quite like the product, but the people in the queue, no.”

The point: A brand's clientele is part of its offer, and can repel as well as attract.

Watch James Hoffmann, the former world champion barista, make a 15-minute video on V60 pour-over coffee and you will start a sane man and end up importing filter papers from Japan and spending 2,000 dollars on a grinder. Rory calls it benign BS: the faff of preparing coffee probably enhances our enjoyment of it, just as food you have cooked yourself tastes better than something out of the microwave. We should not discard these psychological effects.

“The BS we go through to prepare the coffee probably enhances our enjoyment of the coffee.”

The point: Effort and ritual add real enjoyment, so friction is not always waste.

The earliest version of the IKEA effect is the famous Betty Crocker case. Its cake mix needed only water and sold poorly. A psychologist said there was not enough effort for it to feel like cooking, so the instructions told you to add an egg. Despite adding cost and effort, sales improved, perhaps because it now felt like preparing something for your family rather than cheating. Rory notes it was never tested to academic standards, but the lesson for business is simply that sometimes the counterintuitive approach is better.

“Sometimes the counterintuitive approach might be better.”

The point: Removing all effort can remove meaning, and a little required effort can increase a product's appeal.

Rory recalls research his team once did into bottled water. Everybody in the category was agonising over brand imagery of mountain streams and purity, but when buyers were asked, the principal differentiator for them turned out to be whether the bottle fitted the cup holder in their car. That will fit in the cup holder was the deciding thought.

“The principal differentiator was whether it fit in your cup holder in your car.”

The point: Mundane practical fit often drives choice more than the brand imagery marketers obsess over.

The Angel Hotel in Abergavenny, Sunday Times food hotel of the year, did one thing Rory had never experienced. Checking in tired at the weekend, you ask the latest you can have breakfast, braced for: since it's Saturday, we'll let you lie in until eight. Instead: 11 o'clock. From that moment, Rory thought, these are my people.

“From that moment on, I go: okay, these are my people.”

The point: Beating a universally disappointing norm earns instant loyalty.

In a pigeon experiment, birds pecking a switch for three pellets would eat enough and wander off. When the pellets ran low, the dispenser was tweaked to give zero, one, two or three at random, and the pigeons went bonkers, pecking furiously like gamblers at a Vegas fruit machine. That's variable reinforcement: unpredictable rewards are more addictive than constant ones. Rory's theory of bubble tea is that you never know how many blobs will come up the straw, so it's a gustatory fruit machine.

“What you effectively created was a pigeon fruit machine.”

The point: Unpredictable rewards make a product more compelling than reliable ones.

Drawing on Daniel Kahneman's idea that the brain swaps a hard question for an easy one, Rory says that when buying a second-hand car or a house most of us cannot assess the thing itself, so we ask whether we trust the person selling it. That is not irrational: we have half a million years of evolved skill at spotting untrustworthy people and none at evaluating cars. Likeability becomes a proxy, and a salesperson who would feel shame at a dodgy recommendation is a reasonable bet when you cannot research every coffee machine.

“We instinctively choose to judge the personal because it is a lot easier and we are better at it than judging the technical.”

The point: Customers substitute an easy proxy question (do I trust this person?) for a hard technical one, and that substitution is evolutionarily sensible.

Dan Ariely told Rory that drugs for common conditions like asthma sometimes also work against rare cancers. But when prescribed for cancer, patients read the side effects, may cause drowsiness, and conclude it can't be a serious cancer drug. So they exaggerate the side effects, may cause cardiac arrest and worse, and people think that's really unpleasant, it must be a good cancer drug. When you solve big problems in simple ways, nobody believes you, because people expect the downside to match the achievement.

“No one believes you, because they believe the scale of the downside has to be commensurate with the scale of the achievement.”

The point: People distrust cheap or gentle solutions to big problems, which is why small interventions are hard to sell.

Electric cars would see more adoption if the government had not been so eager to push them, Rory says, since they are intrinsically better. But some car people are like those in religion who care little about God but are obsessed with chasubles, vestments and liturgy: more interested in the accoutrements than the purpose. They love engine design and V8s; the means interest them more than the ends. He sympathises, since he would die in a ditch for the beauty of a steam locomotive.

“They're almost more interested in the accoutrements than they are in the core purpose of the religion.”

The point: Enthusiasts value the mechanism, so arguments about function will not move them.

Drinks are fiercely context-dependent. Pernod is enjoyable in France and repellent in the UK; rosé tastes better by the sea; winter Pimm's never worked. Someone drinking a banana liqueur on a Caribbean beach bought the UK import rights, shipped a case home and invited friends to try it in a basement in Fulham. They were retching all over the place, because what tastes great on a Caribbean beach doesn't taste great in Fulham on a rainy October day.

“What tastes great on a Caribbean beach doesn't taste great in Fulham on a rainy October day.”

The point: Taste is partly the setting, so a product can fail when moved out of its context.

Rory says Britain is gloriously messy, stuck mid-Atlantic on measurement: miles on road signs, yards and metres used interchangeably, cheese bought by the pound but weighed in grams. For temperature, Britons tend to use Celsius when it is cold, because a minus figure sounds dramatic, and Fahrenheit when it is hot, because you can say temperatures soared into the 90s. Journalism picks whichever system exaggerates the story it wants to tell.

“Depending on what you want to exaggerate, you choose one or other of the two systems.”

The point: The choice of scale is a framing device that changes how dramatic the same reality feels.

Because humans decide contextually, an easy way to innovate is not to change everybody's mind but to change the context in which they decide. Advertising always wants to change minds, one at a time. Rory argues that if you change the context or frame of reference within which people make a decision, they will make a different decision, and you only have to do it once.

“You can change a million minds or just change one context.”

The point: Redesigning the decision context is cheaper and more powerful than persuading individuals.

Rory sums up his work in one idea: we do not perceive objective reality; what we perceive is heavily mediated by context, and if you change the context you change the experience even if you have not changed the reality. The same chocolate can be perceived, and reacted to, completely differently with a different story or comparison. Put bluntly, from a business point of view, changing context is often a hell of a lot cheaper than changing reality.

“Changing context is often a hell of a lot cheaper than changing reality.”

The point: Businesses can create value more cheaply by altering context than by altering the product.

If Rory ran British Airways he'd shave money off lunch and dinner, which Brits aren't very good at, and hit tea and breakfast, which we do brilliantly, out of the park: tea in a pot in business class. Optimise for surprise. Virgin Atlantic handed out choc ices during the in-flight film, and that's what everybody remembers. Americans were baffled, because the choc-ice-at-the-cinema connection is purely British, just as Brits miss why Americans find Hendrick's cucumber funny.

The point: A cheap, surprising, culturally resonant touch is remembered far longer than expensive competence.

Rory met the man who coined the word 'podcast' very early and thought it might work for American drive time but not much in the UK. He gets a lot wrong, he admits, like dismissing kombucha. The weird thing: give people a 90-minute drive with all the world's music available and they choose to listen to people talking. He would have predicted otherwise. Audiobooks too: his grandmother had a clunky talking-books-for-the-blind device in the 1980s, so why did audiobooks take so long to take off?

“In your car you have access to all the world's music, and yet you'll choose to listen to some people talking.”

The point: People's real preferences often contradict confident predictions, and viable ideas can sit unexploited for decades.

Rory's local paper covers Westerham, where Churchill's home Chartwell lies. A man sold on eBay a coat Churchill had given him as a young gardener, and the paper described Churchill as the former Westerham resident and wartime Prime Minister. Rory realised no other publication in the world would put those words in that order. What is significant about Winston Churchill? He used to live in Westerham, apparently.

“There is no other publication in the world where you would put those words in that order.”

The point: What counts as the most important fact depends entirely on the audience and context.

Abraham Lincoln asked readers to imagine going to church wearing their wife's bonnet: no less comfortable than a top hat, yet agonising unless you could say it was for a bet. Social awkwardness is an unbelievably strong driver, perhaps even physically: one paracetamol reportedly eases social anxiety. Cider is as nice as beer, but beer, the social norm, outsells it seven to one. Guinness benefits from being obvious in your hand, with tricks like 'splitting the G', while cider without ice looks like lager.

“There are things we won't do because we don't want to stand out.”

The point: Visibility and social norms shape consumption as much as taste does.

Rory recommends The Score by C. Thi Nguyen [?], a philosopher at the University of Utah. A rock climber, Nguyen realised that because climbs are ranked numerically by difficulty, he was endlessly chasing higher grades when he should have been enjoying climbs at which he was genuinely proficient. The book examines the difference between the quantification of life and a life well lived: what you measure is what you do.

“What you measure is what you do.”

The point: Numerical scores can hijack an activity, replacing its real value with the pursuit of the metric.

Les Binet recalls asking Rory, at an IPA seminar during his presidency themed on behavioural economics, whether there was an underlying theory behind the anecdotes; Rory suggested prospect theory, which led Binet to Kahneman. Asked today, Rory says the central lesson would be that context has a powerful but often invisible effect on our decision-making.

“Context has a powerful but often invisible effect on our decision making.”

The point: Behaviour changes with context in ways people rarely notice, which is the core lever of behavioural science.

Rory says the English edition of Dawkins's The God Delusion should have been retitled The Football Delusion. Nobody in Kent fears being hijacked by suicidal Methodists or pistol-whipped by Quakers. Football, by contrast, is statistically the impurest of sports, 'pinball on grass', yet its commentary attributes every outcome to human agency. A single lost match brings calls for the manager to resign, and the only non-human agent ever credited is the crossbar that 'foiled Drogba's efforts'.

“If you choose to derive a large amount of pleasure from watching a series of completely random events play out, effectively pinball on grass...”

The point: We construct narratives of intentional human agency around events that are largely random.

Some dentists no longer hand patients an appointment card with the next date filled in. Instead they hand over the card and a pen and get the patient to write the date down themselves. They find the likelihood of people turning up is much greater.

The point: Self-generated commitments are kept more reliably than ones handed to us.

Everyone knows the line: the factory makes quarter-inch drills, the customer buys quarter-inch holes. Rory loves a J. Walter Thompson debate about Black & Decker, involving a copywriter called Thomas, son of the poet Dylan Thomas. Do you perceive a need for a quarter-inch hole and buy a drill? Or do you see a drill that looks really cool, buy it, and go home looking for an excuse to make quarter-inch holes? Rory thinks both are probably true.

“You buy it and then go home and find an excuse to make quarter-inch holes in order to use your drill.”

The point: Desire can precede need; products can create the uses that justify them.

Meaning is highly context-dependent. Normally giving money is called an incentive or a wage and encourages people. But don't pay your wife for sex: Rory jokes he tried it once and four months later got some, so if it works it works very slowly. And don't go to dinner with friends and leave money on the table; take a bottle of wine instead. In that context money is a really bad thing.

“Don't go to a meal with some friends and leave money on the table.”

The point: Money can destroy the social meaning of an exchange that a gift preserves.

Growing up on the Welsh borders, Rory says, he would sooner have severed his own genitals than not have a car, which would have hurt his social life less. He sees a problem for the motor industry: if you drive frequently, driving becomes System 1, automatic and one of life's great joys, especially in the country. Young people who only hire a car for a week in Lanzarote find driving terrifying, so they never discover the pleasure.

“If you only drive for seven days or 14 days a year, driving is terrifying.”

The point: Familiarity turns effortful activities into pleasures, so infrequent use breeds aversion.

Rory notes a potential feature of driverless cars that humans may never cope with. In theory you would not need traffic lights: with enough precision, cars could criss-cross a junction at 70 mph, rumbling through alternately about three feet apart and judging the gaps perfectly. It would hugely increase the efficiency of the road network, but people would simply find it too terrifying.

The point: The engineering optimum can be psychologically unacceptable, so human comfort caps technical efficiency.

Rory finds Americans very absolutist and rule-based, the stuff of Larry David. In Los Angeles he unwittingly drove into a strip-mall car park exit and met a car leaving. Reversing onto a fast highway was too risky; the other driver only had to reverse ten feet. In Britain there'd be twenty seconds of bartering: hands up, 'my bad', a grudging reverse. His sister-in-law warned it would take a minute and a half, because something unusual had happened and the rules had been broken. Sure enough, the driver got irate before reversing. Brits, by contrast, shrug at naked Germans on a Fuerteventura beach.

“In the UK there's a sort of 20 second period of bargaining where you put up your hands.”

The point: Rule-bound cultures handle exceptions badly, while informal negotiation resolves them quickly.

Estate agents go to great lengths to make sure the vendor and buyer of a house don't meet until a survey is done and documents signed, because if one dislikes the other personally, they won't sell or won't buy. When Rory told this story, someone said it happened to their mother: the agent reported an offer at the asking price, and she asked, 'It wasn't those shitty people who came round yesterday, was it? Well, I'm not selling.'

“It wasn't those shitty people who came round yesterday, was it? Well, I'm not selling.”

The point: Personal feelings can override large financial interests, so the human element still decides big transactions.

Rory's generalisation: every number one place is overrated and every number two is underrated. Fame confers itself winner-takes-all, as with Salcombe in Devon, too popular for its own good. As with brand leaders, the most famous is in more people's repertoires and bought more often by them, so it gains disproportionately. It might be 5 per cent better than the next, and 15 per cent worse when crowded. Go to Umbria not Tuscany, the Lot not the Dordogne, Chicago, the best big American city, not New York. You are in the right place if half the breakfast room are not tourists.

“Every number two place is underrated and every number one place is overrated.”

The point: Popularity compounds beyond merit, so the runner-up often offers better value.

Taleb's lesson is that a brilliant mathematician using the wrong maths is dangerous: a whole edifice can rest on one wrong assumption. In food, the wrong assumption is trusting what people claim. In research groups, people want to seem responsible and middle class, so they ask for healthy options, then do not buy them. Sometimes fast food is a cheap hot lunch, sometimes a treat, and a salad is not a treat. The McDonald's salad still helps: parents who do not want a burger can take the kids.

“If you research people, they will always say, I want a healthy option. In the event, they don't buy the healthy option.”

The point: Stated preferences in research reflect self-image, not purchasing behaviour.

Richard Thaler said a nudge is acceptable if it leads to a behaviour the person doesn't later regret. Rory looks for a stronger category, which he calls 'reverse regretters' [?]: behaviours people wish they'd adopted earlier. Every elderly person who gets a stairlift says 'I wish I'd done this a year ago', as his late father did. His advice: if you've contemplated one for five seconds, just get one. When people wish they'd started sooner, the barrier is obviously psychological, not price or product.

“If you've even contemplated getting a chairlift for five seconds, just go and effing well get one.”

The point: Behaviours people regret not adopting earlier reveal psychological barriers ripe for behavioural solutions.

Ogilvy sent a credit card offer in envelopes marked final reminder in red, because the deal was about to expire. The team thought it cute. A significant minority went bananas, complaining that their postman now thought they did not pay their bills. To Londoners this was incomprehensible: in a big city you hardly know your postman and do not fear gossip. In a small village the postman drinks in the same pub as your friends. Nobody on the team had considered it because Londoners would not be bothered.

“My postman thinks I don't pay my bills.”

The point: Teams miss how a message lands in contexts unlike their own, such as small communities.

All maps distort, and when everyone uses one for years the distortions shape behaviour. The Central line is overused because it is red and horizontal; the Victoria line looks curved, so is underused. Fulham residents think they live centrally because they're on the Tube. Friends moved from Fulham to Herne Hill for an extra bedroom, expecting the wilds, and found their commute halved. The map also hides that Paddington is an easy downhill walk to Lancaster Gate. Rory's tip: ask what isn't on the map.

“Get a copy of the Tube map and ask yourself what isn't on it.”

The point: Everyone relying on the same model creates bargains wherever that model is wrong, which is the problem with economics.

Frederick the Great wanted Prussians to grow potatoes: two carbohydrate crops meant less bread price volatility and lower famine risk. But potatoes looked disgusting, peasants said even the dogs would not eat them, and compulsion failed, with some people reportedly executed for refusing. Plan B was marketing. He declared the potato a royal vegetable, planted a royal patch and had it guarded day and night, with secret orders not to guard it very well. Peasants knew that anything worth guarding is worth stealing, and soon there was a massive underground potato-growing operation.

“If something is worth guarding, it's worth stealing.”

The point: Making something look scarce and exclusive can achieve what compulsion cannot.

When Rory was young, the goal in life was a bigger telly. In a Carry On anecdote, Kenneth Williams's mother hears a neighbour has bought a 14-inch console set and quips that 14 inches would be enough to console anyone. Screens kept growing, and a cheap monitor still transforms productivity. Yet his children watch films lit by Roger Deakins on a phone the size of a letterbox. He also notices great films seem poor on a plane and poor films seem great, so the viewing context reshapes the judgement.

The point: The medium and setting change how good an experience feels, independent of its content.

Marketing is more powerful than even marketers realise because so much value is perceptually determined. Sell a functional drink like Trip in a big can and it probably would not have worked. Anything medicinal or perceived to be potent must be small. The extreme case is ginger shots, at about two pounds fifteen: they are expensive because they are small. Proceed as if consumers are rational utility-maximisers with an objective view of reality and you will get a lot badly wrong, because most really interesting ideas have an element of the nonsensical.

“They're expensive because they're small.”

The point: Small size and high price signal potency in functional products.

Much of our physiology is not directly controllable: you can't will your heart rate up or your pupils to contract, though you can hack them obliquely by walking into a dark room. Rory likens it to an automatic gearbox. People who drive manuals say they'd hate the lack of control, but drive one long enough and you learn to control the gear it's in, by easing off the accelerator before a hill, not with a lever.

“You kind of control it through oblique means, not direct means.”

The point: Systems we can't command directly, bodies or people, can often be steered indirectly.

Mark Earls, author of 'Herd', pointed out that some products are immune to herd effects, with haemorrhoid cream top of the list. Apps are often used privately too, but you can still amplify social proof: the difference between a million and 100,000 users barely registers, while the difference between 407 and 14,000 reviews matters a lot. You can also build in gratuitous visibility, Seth Godin's 'Purple Cow': Monzo made its card coral, so it gets noticed even though most payments are on a phone.

The point: Social proof works in thresholds, and products used privately need deliberately added visibility.

Rory opens his talk by declaring there is a natural solution to everything. The answer to the male obesity crisis is simple: Hawaiian shirts, because the bigger you are, the better you look in one. He jokes it is no surprise Tom Selleck was so keen to play Magnum PI. The gag sets up his theme that changing the context or presentation of something can change how it is perceived without changing the thing itself.

“The solution to the obesity crisis for men: Hawaiian shirts, the bigger you are the better you look.”

The point: Reframing presentation can turn a perceived flaw into an asset.

Explaining why gender is less of a flashpoint in Britain, Rory notes the UK has had pantomime for 200 years, men dressing as women and women as young boys, plus Shakespeare's cross-dressing. Some Like It Hot was highly controversial with American studios because men dressed as women. His grandmother, a conservative woman in provincial Wales, went to see it, and it would never have occurred to her there was anything controversial about it. It was simply funny.

“It would not have occurred to her literally that there was anything controversial about this at all.”

The point: Cultural context determines whether the same content reads as harmless or provocative.

In a self-confessed rant, Rory argues that possessiveness over kitchen worktops is an obstacle to UK economic growth: try buying an air fryer and an ice cream machine and every inch is fought over. He asked someone from House & Garden why the people in the magazine had spent £18,000 on a kitchen yet owned no microwave or TV. They said they take all that stuff out before the photo shoot. So the whole world labours under the delusion that a proper kitchen contains nothing with a plug attached.

“The entire world is labouring under the delusion that a proper kitchen contains nothing with a plug attached.”

The point: Aspirational imagery quietly sets norms that suppress useful consumption.

Daniel Kahneman emailed Rory and a handful of Nobel laureates a friend's problem: an Austin farmers' market seller with healthy margins, selling lumpy vegetables to hipsters, could never quite sell out. Scarcity works for identical goods, but we assume the last cucumber is a bit manky. Richard Thaler, being in Chicago, reached for price; others for scarcity. Rory suggested something different: at the end of the day the seller loses interest, starts packing the van, the table is half covered, and it no longer looks as if you are supposed to buy.

“Maybe it's something completely different.”

The point: Before reaching for price or a named bias, look for simple contextual cues that are actually putting people off.

Take the clue 'Does perhaps rush around (4)'. The answer is 'deer', which seems insane until you see that 'does' is the plural of doe and 'rush' is a reed, spelled backwards. The surface meaning misleads you into reading both as verbs. Rory argues much human behaviour works the same way: there is a plausible surface explanation and a deeper hidden purpose, and you only crack it once you stop taking the clue literally.

“Human behaviour is an enigma. Learn to crack the code.”

The point: Look past the stated reason for a behaviour to the cryptic one underneath.

One of the strangest things about human behaviour, Rory says, is how little ice cream we eat when nearly everyone loves it: there are few social occasions where it is permitted, and bringing one into the workplace would seem weirdly childish. Stranger still, Europe's biggest ice cream consumers are in places like Russia and Sweden, not Italy. What drives consumption is not temperature but sunshine: when the sun breaks through briefly, Swedes strip off, sunbathe half-naked in Stockholm and buy an ice cream.

“The thing that drives ice cream consumption isn't temperature, it's sunshine.”

The point: Consumption is governed by social permission and cues, not by the obvious functional driver.

Near where Rory grew up, close to Pontypool, many people had retired from ICI's nylon factory and its R&D unit. They lamented that they had developed remarkable fabrics, ones that stayed clean, kept you cool, even artificial silk, and then everybody went and wore jeans. They had asked what consumers wanted logically; what consumers wanted from trousers was a weird psychological connection with the American West. Rory, who owns no jeans and finds them uncomfortable and hard to maintain, says: go figure.

“We developed tons and tons of remarkable fabrics, and then everybody went and wore jeans.”

The point: Products win on psychological meaning, which logical R&D cannot anticipate.

Rory discovered that in the US a speed limit is not just a restriction but advice: a 60 sign means it is safe to drive at 60. In Britain, the national speed limit sign leaving a Kent village means 'use your own judgement; it is not illegal to hit 60 where conditions allow.' So Americans ask if Brits are insane to drive winding lanes at 60, and Rory's American-programmed Mustang Mach-E sees a 60 sign and immediately accelerates to 60.

“In the United States, that 60 sign means it is safe from now on to drive at 60 mph.”

The point: The same sign carries different meanings in different cultures, changing behaviour.

The Tube has its own social rules. Rory recalls the Hungarian humorist George Mikes, whose routine of advice for foreigners in Britain deliberately gave inappropriate tips, including: on entering a London Underground carriage, be sure to introduce yourself to all the passengers. In Australia that might be fine on a long journey; in Britain people would assume you were a serial killer, though northerners might chat.

“On entering a London Underground carriage, be sure to introduce yourself to all of the passengers.”

The point: Social norms vary sharply by place, and violating unwritten rules marks you as a threat.

On the Italian rule against cappuccino after about 11am, Rory warns not to let Italians tell you it is because they are sophisticated: a large number of them are lactose intolerant. And while we admire them for drinking an unbelievably sour espresso, nearly all of them have dissolved about three sugars in it.

“Don't allow the Italians to tell you that's because they're sophisticated.”

The point: Customs that look like refined taste often have mundane practical origins.

The advertising man Jeremy Bullmore observed that if you turn up at a black tie event in a suit because you cannot afford a dinner jacket, that is a compromise and you feel dreadful. If you turn up in a suit because you have a dinner jacket at home but cannot be bothered or object to wearing it, that is a choice and you feel absolutely fine. The product is identical; the meaning, the sweeping of the floor, is completely different.

“There's a huge difference between a choice and a compromise.”

The point: The same behaviour feels good or bad depending on whether it reads as chosen or forced.

Going independent means work and leisure stop being mutually exclusive. Rory spent about eight days in Tenerife recently, three of them working. Daniel Kahneman even argued you should work a couple of days in the middle of a holiday, because constant leisure loses its novelty: loll by the pool for two days, do a day's work, and the pool feels more enjoyable again. Work in the morning, holiday in the afternoon is a fantastic combination the conventional world of work never allowed.

“They're not mutually exclusive, they're combinatorial.”

The point: Interrupting a pleasure renews it, so mixing work and leisure can increase enjoyment of both.

Rory gives a classic example of salience bias. British people probably assume that because koalas and kangaroos are equally famous, there are similar numbers of each. In fact kangaroos are insanely common, hopping all over the countryside, while most people, including his Australian host, have only ever seen a koala in a zoo.

The point: Fame and salience distort our sense of how common things are.

Logical schemas are modelled on simple physics and children's maths: solvable, with a single right answer. By their rules there can be no magic. Logic says that to improve people's experience of a hotel you must improve the hotel itself, not their perception of it. That is wrong-headed: context is a marketing super weapon precisely because it works magically. The fact that Apple, the first highly subjective technology company, is worth around a trillion dollars should prove the point.

“Context is a marketing super weapon, and it works because it works magically.”

The point: If you assume people perceive objectively, you confine yourself to expensive objective improvements and miss cheaper psychological ones.

Rory tells the audience the reframing trick is being played on them right now. There is dry ice, the most impressive screen backdrop he has presented in front of, magnificent lighting, cubes and fantastic sofas. Then he tells them to look at the roof: they are in a shed and had not noticed. Without all the staging they would feel like they were sitting in a prison.

“Just look above your head. You're actually in a shed.”

The point: Value is a product of how we look at something, not only of what it is.

Status is a relative currency, as the 1930s joke has it: a rich man is anybody who earns more than his wife's sister's husband. Rory finds that sad. Had you visited Louis XIV with a 55-inch LED TV, he would have given you half of France for it. But because everybody has one, we don't regard it as magical. We derive less pleasure from amazing things once they become widely available.

“If you'd gone to Louis XIV with a 55-inch LED TV, he would have given you half of France for that.”

The point: We value things relative to others, so abundance erodes the pleasure of genuine marvels.

We believe drugs are effective in proportion to how odd they taste. Nobody in oncology wants to be asked 'strawberry or blackcurrant?' Rory never trusted Nurofen Meltlets because they were too tasty. Health food tastes weird: drink wheatgrass and you think it must be doing you good, though you could lick the underside of your Flymo with much the same effect. A tonic wine, Sanatogen [?], reportedly added its last ingredient purely so it wouldn't taste too nice.

“You can lick the underside of your Flymo and have much the same effect.”

The point: Unpleasantness is read as a signal of efficacy.

Half joking, Rory questions the assumption that men dislike marriage, noting that married men tend to live longer and happier lives. Perhaps what puts them off is the wedding: days spent debating flowers and coloured sashes. Picture instead a wedding redesigned as a few days of travel and passion, with friends buying you premium electronics, and men would be begging reluctant girlfriends to commit. He adds that after the wedding nobody buys the husband a present again, or if they do it is a tablecloth.

The point: Immediacy bias means the up-front experience of a decision can outweigh its long-term rewards, so redesigning the entry point changes uptake.

A colleague who worked on Marks and Spencer for years always wanted, but never dared, to tell them he would not work on any advertising until they sorted out the flooring and lighting in the clothing department. Rory suspects one procurement process buys flooring and lighting for both food and clothing. You cannot get someone to spend 250 pounds on clothes and shoes under industrial lighting on an easy-clean floor. He jokes there is a rule that spending over 100 pounds entitles you to a rope-handled bag; a polythene bag after spending 200 feels wrong.

“You can't get someone to spend 250 quid on clothes and shoes if you have basically industrial lighting and a bloody easy-clean floor.”

The point: The environment must match the price point, or it silently undermines the value of good products.

Rory complains that there is no vomit on London's streets any more. Vomit on the street is bad in and of itself, but statistically it is a tail indicator of a hundred other people having a good time who are not being sick. So if nobody is being sick in the street, the invisible downside may be that far fewer people are having a good time.

“It's a tail indicator of a hundred other people having a good time who aren't being sick.”

The point: Eliminating a visible harm can hide a much larger invisible loss.

Very few people will say 'I bought an electric car because two of my friends have one'. They will cite the environment or cost savings. Yet, Rory says, the single biggest determinant of whether people buy an electric car is whether they already know other people who own one. Because customers misrepresent their own reasons, businesses misdirect huge effort.

The point: Social proof drives decisions that people later explain with rational reasons.

Quoting Michael McIntyre, Rory notes that wine is the only thing a restaurant asks you to smell to check it is not off: nobody brings you the milk for your coffee and asks you to check it. Friends in the restaurant trade believe most people pretend to like wine. Serve a corked bottle and diners nearly always drink it. The only clue is that they do not order a second, because nobody is confident enough to say something is wrong.

“Even if you serve a bottle of wine that's fundamentally corked, people will nearly always drink it.”

The point: Fear of social embarrassment makes people pretend to enjoy things and hides product failure.

Asked for the most important lesson of his life's work, Daniel Kahneman didn't cite prospect theory; he said nothing is as important as you think it is while you are thinking about it. Rory applies it: a diner's menu makes choice salient, so choice seems to matter; McDonald's makes speed salient. The EasyJet pilot moved attention from 'no gate' to 'no walk'. Boutique hotels did the same, dropping 1970s opulence and putting an espresso machine in the room.

“Nothing is as important as you think it is while you are thinking about it.”

The point: Whatever you make salient is what people judge you on, so choose the focus.

Engineers, doctors and scientists are obsessed with solving the problems of reality. But once a society reaches a basic level of wealth, Rory says, most problems are actually problems of perception. So the question becomes how many of life's problems can be solved by tinkering with perception, rather than the tedious, hard-working and messy business of trying to change reality.

“Most problems, once you reach a basic level of wealth in society, are actually problems of perception.”

The point: Changing how something is perceived is often cheaper and more effective than changing the thing itself.

New product design goes to engineers, who want to impress other engineers and be superior on engineering dimensions. But we don't perceive reality objectively. Engineers over-optimise and miss the law of diminishing returns: consumers really notice if something is ten times better, but barely care if it is 1.3 times better. Many businesses optimise for objective reality rather than perception. Rory's advice: don't try to change the objective reality of something, change how it is experienced.

“Optimise for perception, not for reality.”

The point: Investment should target what people perceive, since objective gains below a perceptual threshold earn nothing.

Rory is clear that ordering screens are great: queues are shorter and many people prefer them. They also increase the number of men ordering a meal with two burgers in it, because there is no social embarrassment of asking a person who can tell you are eating alone.

“You don't have the social embarrassment of requesting that from someone who can tell you're a single customer.”

The point: Removing a human witness changes what people feel able to choose.

A friend who is warden of an Oxford college told Rory that tourist visits have completely bifurcated: some colleges swarm with tourists, others get nobody. The entire discriminator is whether the college appeared in a Harry Potter film. New College has a tree that featured in one, and tourists swarm in. Next door, an architecturally magnificent college that never appeared gets tumbleweed. Likewise, a replica Harry Potter train would bring people to London just to travel to Manchester.

“The entire discriminator between those two is whether the college features in a Harry Potter film.”

The point: Fame and story, not intrinsic quality, concentrate attention and demand.

Rory thinks Spanish cooking has performed spectacularly in the last 20 years, arguably rivalling France and Italy, but considers paella a failed biryani. Authentic Valencian paella is made with rabbit and vegetables; the chicken, prawn and chorizo version is a Tex-Mex-style adaptation that Valencians regard as we might a Hawaiian pizza. Similarly, someone found that many Italian classics, supposedly handed down by generations of grandmothers, really date from the 1940s and 50s.

“Paella is actually a failed biryani.”

The point: Supposedly ancient traditions are often recent, mass-market adaptations.

Talking with the behavioural economist George Loewenstein, Rory noted that solar panels are sold on the line it should pay for itself in ten years. That sounds like shelling out a fortune and, a decade later, just getting your money back. Framed as an investment, a 10 per cent annual return, it is far better than almost anything else available. Nobody sells a 10 per cent investment product as get your money back in ten years.

“Nobody sells a 10% investment product on the 'get your money back in ten years'.”

The point: The same economics framed as an investment rather than a grudging outlay changes how attractive it feels.

Rory disputes the view of marketing as fairy dust sprinkled on a product's real, intrinsic value. We value things according to what they mean, not what they are, and meaning depends on context, so storytelling, framing and recontextualisation create real value in the mind just as a factory does. He adds, admitting it is ambitious, that perceived value is environmentally friendly: imbuing a product with meaning uses far less carbon than making it three times bigger or five times faster. Psychological moonshots are easier than technological ones: a train ten times faster is near impossible now, ten times more enjoyable is doable.

“Value can be created in the mind every bit as much as it can be created in the factory.”

The point: Creating meaning is a legitimate, cheaper and lower-carbon route to value than physical improvement.

Everyone was convinced Pizza Express pizzas had got smaller, which the company denied, saying it still used the original dies. Rory suspects it is because everyone else's pizzas got bigger: Domino's and others started making enormous ones, so by comparison Pizza Express looked small, when they had seemed surprisingly big at launch. He adds that he hopes Britain does not become like America, where what matters about a restaurant is how happening it is rather than how good the food is.

The point: Perceived size and value are relative to the surrounding market, not absolute.

Picking up an idea from Tyler Cowen's Marginal Revolution blog, Rory jokes that education does not work by teaching you things. It works by giving you the impression you have had a very good education, which gives you an insane sense of unwarranted self-confidence, which then makes you very successful in later life. "So, welcome to Oxford, ladies and gentlemen."

“It actually works by giving you the impression that you've had a very good education, which gives you an insane sense of unwarranted self-confidence.”

The point: Some of the value of institutions comes from the confidence and perception they create, not their content.

There are no purple photons: purple is not on the light spectrum. When you see red and blue, the brain should average them to green, but the green sensor in the middle detects nothing. Rather than declaring a system error, the brain invents purple, a colour that essentially means 'not green'. Rory's worry about big data is that it assumes a neat mapping from objective reality to behaviour, when perception can make identical things look different, as with the dress, and create things that exist only in the head.

“It's a colour to represent the weird unexplained absence of green.”

The point: Behaviour runs on perceptions the brain constructs, which may not correspond to anything objectively real.

Rory watches podcasts on YouTube partly because his TV has his best sound system, but mainly because they don't demand full attention. In fairness to tea, coffee and tobacco, he jokes, they are three drugs you can consume while performing other useful functions, which is not generally true of acid. Likewise radio is the medium of the industrious, because you can listen while getting on with something else. Video podcasts give scene-setting you glance up at, without the undivided attention television demands.

“Tea, coffee and tobacco are three drugs you can consume while performing other useful functions.”

The point: Products that coexist with other activities fit into life more easily than those demanding full attention.

Humans must both stand out and fit in, and brands and products tread the same line, captured in Raymond Loewy's MAYA principle: most advanced yet acceptable. Loewy designed a fan so quiet that people assumed it was not working. Someone told Rory they were using AI to design a silent electric razor, and he objected that the crackle and whine is how he knows it works. Digital cameras need a click and electric cars get artificial hum. A Top Gear presenter found a near-silent Rolls-Royce Spectre so quiet that a faint air-conditioning noise drove him nuts.

“You can be too weird to succeed.”

The point: Innovations need familiar cues to be accepted, and removing a cue such as noise can make a product seem broken.

Rory reports that somewhere in New Mexico, instead of a maximum speed, roads have a recommended speed of 50 to 70 mph, and people tend to travel at about 60. With a maximum of 70, people travel at 75. The way you describe something has a big effect on behaviour, because context determines meaning, meaning determines emotion and emotion determines behaviour.

“It's context that determines meaning and it's meaning that determines emotion and it's emotion that determines behaviour.”

The point: Framing a limit as a recommended range changes behaviour more than a hard ceiling.

Rory cites Jeremy Bullmore, former JWT creative director, who recalled that in his childhood all cheddar came with a rind. Then someone sold rindless cheddar at a premium as a convenience: no paying for rind, no cutting it off. Some 25 to 30 years later, once memories had faded, producers launched farmhouse artisan cheddar with the rind left on, also at a premium. Rory calls this recursive fashion, one of the curses of capitalism: human neophilia means what is different grabs attention, and when old things return the context gives them new meaning.

“One of the curses of capitalism is recursive fashion.”

The point: Novelty can come from reviving the old, because meaning depends on context and memories are short.

When the Rush hairdressing chain opens a salon it spends £30,000 on the toilets. Most customers never use them, but the 10 per cent who do expect a mop bucket and stacked hair products, like walking through the 'staff only' door of a five-star hotel. Rush gives them gold taps instead: optimising for surprise, not predictability. Buc-ee's in Texas and a Turkish petrol station chain also make the toilets the hero. Since the brain predicts and attends to what's unexpected (Andy Clark's The Experience Machine), emotional effect comes from the DoubleTree cookie, not from meeting service levels.

“We're optimising for surprise. We're not optimising for predictability.”

The point: Delight comes from exceeding expectations in neglected places, not from uniform competence.

People call Ryanair's survival a mystery, since everyone affects to despise it. Rory says its genius is that the real experience is always slightly better than you expect. The night before, you're terrified you'll be fined 70 pounds because your cabin bag is a millimetre too large. Then it's kind of all right, and you land thinking: I wasn't pistol-whipped by Serbian mercenaries, so I'll chalk that up as a win.

“I wasn't actually pistol-whipped by Serbian mercenaries, so I'll chalk that up as a win.”

The point: Satisfaction is measured against expectation, so low expectations met are a win.

Guidara told his sommelier: on a wine pairing, don't spend $20 on each of five wines as most restaurants do. Give them a $15 wine, then another, and use the saving to spend $50 or $60 on the final one, perhaps a grand cru Burgundy or a cracking dessert wine. It's the peak-end rule: don't do everything quite well; have one standout moment and put it at the end, at no extra cost. Rory links it to Gary Stevenson's point that economic models rely on a single representative agent, treating the average as the whole. Don't optimise for average.

“Don't optimise for average.”

The point: Reallocating the same budget into one memorable peak, placed at the end, transforms the remembered experience.

Rory sees more 'defensive opinion forming' now: as a child everyone admitted wanting an Aston Martin but couldn't afford one; now young city dwellers without one pretend they don't like cars. Aesop's sour grapes is pretending you never wanted what you can't have. Its counterpart is sweet lemons: something bad happened, so you reframe it as positive, 'my six years in Leavenworth were a really formative experience'. Bad things you had to have become positives; good things you couldn't have become negatives. Aesop, 2,500 years ago, was the first behavioural scientist.

“My six years in Leavenworth prison were a really formative experience.”

The point: People reshape their preferences to fit their circumstances, so stated attitudes can be self-protective rationalisations.

Users complained about Spotify's buffering, so managers asked the engineers how low they could get it. They came back with about 220 milliseconds. Asked whether they could do better, they said yes, but there was no point: to the human brain 220 milliseconds is near enough instantaneous, while 250 is noticeable. As with a Retina screen, past a certain threshold we simply don't notice.

“Making objective reality better than our ability to perceive it is a really dumb effort.”

The point: Optimise to the threshold of perception, not beyond it.

When Ogilvy moved to Blackfriars, Rory asked staff about their commute. Several who came into St Pancras said it was difficult, and he thought it absurd: take the Thameslink from the front of the train and you are fifty yards from the office. Then he realised Thameslink, like other railways, is not on the Tube map. North Londoners, like the finance function, barely know South London's rail network exists, and imagine South Londoners carrying their possessions in a spotted handkerchief on a stick.

“Then I realised Thameslink isn't on the Tube map.”

The point: People make decisions using the map they have, so whatever is off the map is invisible.

Sometimes the value lies in the time spent, as with Saturday farmers' markets, which are essentially Tesco reversed: seven stops and a chat with every seller. Starbucks restricted baristas to making no more than two drinks at once, deliberately slowing service, because batch-processing coffee made customers feel they were on a production line rather than getting a handmade drink.

The point: Visible effort and time can be part of the product, so efficiency can destroy value.

Rory teases the Oxford students that they mostly go round on bicycles and therefore become environmentally conscious, even though they think it is the other way round. He offers it as a pointer to a bigger idea worth investigating: that our attitudes are often a product of our actions rather than the cause of them.

“You ride bicycles and therefore become environmentally conscious, even though you think it's the other way around.”

The point: Attitudes often follow behaviour rather than drive it.

The number of people trying to carry water through airport security has not fallen in eight years, despite rising awareness of the rule. Ogilvy's hunch is choice architecture: since low-cost airlines began charging for hold luggage, most frequent flyers own a maximum-size clamshell carry-on, which you cannot open in public without spilling your underwear on the floor. One simple solution, with no money going to Rupert Murdoch: put tables in the run-up to the security lane so people can retrieve the bottle without hideous embarrassment.

“An awful lot of behaviour is driven not by information, not by persuasion, it's just driven by the environment in which people find themselves.”

The point: Many behaviour problems are caused by the environment, not a lack of awareness, so fix the environment.

Rory always thought telephone conferences were disastrous because they were a form of sensory deprivation: with nothing to look at, it was almost impossible for your mind not to wander. One colleague took delivery of a flat-screen television and demanded it be put on the wall during a client conference call. So he was not surprised that video calls turn out to be a reasonable approximation of meeting in person.

“Telephone conferences were a form of sensory deprivation.”

The point: Attention needs something to anchor to, which is why audio-only meetings invite mind-wandering.

Rory argues wine is overrated: there is never a situation where white wine is nicer than a gin and tonic, and the fuss over distinctions is like The Princess and the Pea. It is also inconsistent. If he opened a box of cornflakes to find them soggy and black and Kellogg's said it was the 2016 cornflake with slightly late rainfall, he would not savour the new notes, he would tell them to sort their act out. With wine, total lack of quality control is treated as acceptable, even interesting.

“I want the third bloody box to taste like the first.”

The point: Category conventions decide whether inconsistency reads as a flaw or as a feature.

Tim Harford told Rory about an AI trained brilliantly to recognise animals. When shown pure white noise, it claimed there was an animal there, because it had never been shown anything that wasn't an animal. It had no way of saying 'you're taking the piss, this is just noise', so it picked whichever animal was most plausible. The same confabulation happens in humans, which is why we see faces in clouds. Rory adds that AI's recent leaps have nonetheless made him less sceptical.

“It had no conception of being able to say: you're taking the piss here, this is just a load of crap.”

The point: Systems trained only on positive examples will find patterns in nothing.

Some pain is enjoyable. Put antiseptic on a cut on your shin and you want it to sting, because the sting is evidence of efficacy. The fact it hurts proves it is working, so your reaction to the pain is not entirely discomfort: you have reframed it as proof that it is doing its job.

“The fact that it's hurting proves it's working.”

The point: Unpleasant sensations can be valued as signals of effectiveness.

Aged about 16, around 1979, Rory heard a Jimmy Young radio phone-in about litter, with caller after caller demanding that people who threw litter from cars have their hands chopped off. At the very end, a Mr Foulk [?] rang from Oxford to say he thought litter was a very good thing: everywhere interesting, like New York or Venice, has rubbish all over the place, whereas the Cotswolds have little litter but are as boring as hell. Rory later met him at university: Ray Foulk [?], who organised the Isle of Wight festivals with his brothers.

“If you go to the Cotswolds there isn't much litter but it's as boring as hell.”

The point: An apparent negative can be a sign of the vitality we actually value.

Posh Turkish restaurants have one rule: you cannot sell a doner kebab. Rory once made the perverse argument that the doner kebab may be the world's finest food, because when your brain is drunk and stripped of all its protective inhibitions, it is what you really want to eat. If we are intellectually honest, we have to acknowledge its magical property.

“When your brain is drunk and stripped of all protection, it's what you really want to eat.”

The point: Preferences depend on state and context, so judging food in only one state misses its value.

Rory argues the English are actually good linguists in one respect: they are practised at understanding English spoken oddly by non-native speakers. Get the gender wrong in French, ask for 'la limonade', and nobody understands for six minutes. That might be wilful perversity, he jokes, but since there have been no recorded cases of wilful perversity in France in 200 years, that explanation can be discounted. In America, where people rarely hear unusual accents, he cannot pronounce 'water' in a way anyone understands.

“Since there have been no recorded cases of wilful perversity in France in the last 200 years, I think this explanation can be discounted.”

The point: Exposure, not innate talent, explains much of what we attribute to national character.

A creative team briefed to advertise a small florist in a Gloucestershire town noticed that nobody registers flowers in a five-star hotel because they are expected there. So they took the shop's leftover flowers at the end of each week, made up a few bunches and left them where nobody expects flowers: a phone box, the men's and women's public toilets, a bus shelter, the police station. A small card underneath suggested surprising someone with flowers and gave the shop's details. In a public lavatory, Rory says, a bunch of flowers really lands.

The point: The same thing has far more impact in an unexpected context, so where a message appears can matter as much as the message.

A clever former colleague bought a house with a pool in the Surrey Hills for the price of a flat in Croydon. Asked why it was so cheap, she explained it was full of old-lady furniture because the previous owner had been elderly, as is typical of five-bedroom British homes. Buyers judge the previous occupant's taste from the decor and mark the house down, even if they plan to redecorate anyway. Rory admits it is not totally irrelevant: if you knew Rick Rubin had lived there, you would buy it sight unseen.

“You should ignore the photographs of interior decor when you look at a property.”

The point: Buyers overweight easily changed surface features, which creates bargains for those who see past them.

Good mathematicians know the limits of models; economics has a convincing, internally consistent model that is not true. Ask an economist why Uber succeeded and you get economies of scale. Rory argues most disproportionately successful companies overcame a psychological obstacle to adoption. Apple stopped making computers ugly. The iMac said you could have a computer in a room without turning that room into an office; before, a beige box made it a space for photocopiers and fax machines. No spreadsheet or focus group would have produced that. Apple's coders asked what Steve even did.

“You can have a computer in a room in your house without turning that room into an office.”

The point: Breakthrough success often comes from removing an unmeasured psychological barrier rather than improving specifications.

Noriaki Kano of the University of Tokyo plotted how features translate into satisfaction, and some are negative. TV remotes with dozens of buttons frighten people, some tape over all but volume and channel, so Samsung's six-button remote is welcome. Hotels think taking your bag is service, but after a day guarding your suitcase you hate a stranger grabbing it. A non-leaking milk carton is a table stake: nobody gets excited, but a leak loses you for ever. Airlines once competed on 45 versus 57 per cent seat recline, which nobody cared about: humans are comfortable sitting or lying, nothing in between.

“There are two comfortable positions for humans: sitting in a chair and lying in a bed.”

The point: Features relate to satisfaction non-linearly; some add nothing, some subtract, and only a few delight.

In an experimental philosophy thought experiment, a chief executive is told siting a factory away from the river is better for the environment. He says he doesn't care, put it where it makes most money, and it goes by the river. Did he intentionally harm the environment? 82 per cent say yes. In the mirror case, where the profitable site happens to be away from the river, only about 18 per cent say he intentionally helped the environment. Self-interested motives get blamed for harm but never credited for good.

The point: Our judgements of intention are asymmetric, so profit-driven actions are penalised for bad side effects and ignored for good ones.

For years Rory worked in Soho Square and loved Korean food, and right round the corner was a Korean restaurant. But he never saw anybody in there, so he never went in. Only later did he learn that, perhaps for cultural reasons, they never seated people by the window. The restaurant was half full or full all the time, but everyone sat at the back, leaving the visible window section permanently empty.

“We make these kinds of inferences all the time without being aware of doing so.”

The point: Customers infer quality from visible crowds, so where you seat people is a marketing decision.

Rory used to have an old but fancy Mercedes whose seat squab inflated on one side as you went round a corner, like airbags gripping you. It made driving feel like a mission to Mars, incredibly thrilling without being any faster, so he always chose to drive that car. No advertising campaign ever mentioned it, to his knowledge; in the good old days, someone would have found that USP and banged a drum about it.

“It just made it incredibly thrilling to drive without being any faster.”

The point: Small sensory features can change the experience of a product more than its measurable performance.

At Flat Iron, the restaurant co-founded by Charlie Carroll, diners are given a little token at the end and leave onto the street with a miniature ice cream cone. Rory admits such touches can seem a bit gratuitous, even hokey, but the human brain notices what it wasn't expecting much more acutely than what it was. Combined with the peak-end rule, a surprise at the end colours memory of the whole meal.

“You notice what you weren't expecting much more acutely than things you were.”

The point: A small surprise at the end of an experience punches far above its cost in memory.

In Rory's Kent village, a wood-fired pizza van turns up on Mondays only, alongside a souvlaki stall. He is convinced his children ate pizza from that van much more often than they would have if there had been a pizzeria at the end of the drive. It is the scarcity effect: if we don't buy it now, we'll wait another seven days. The rest of the week the van is somewhere else, and he has no idea where.

“If we don't buy it now, we'll have to wait another seven days.”

The point: Limited availability can increase consumption compared with constant access.

New behaviours form in pockets: you need to create a space where they are normal. Rory's example is naturism. You would not have much fun being the only naturist in Sevenoaks; you would be known as Pervy Dave and nobody would let their children near you. But create a beach and you start making progress. Geography matters for visibly consumed goods, and many products are advertisements for themselves, such as the green flash on electric cars' number plates.

“You wouldn't have much fun being the only naturist in Sevenoaks.”

The point: Unusual behaviours spread from concentrated pockets where they already feel normal, not from scattered individuals.

The Greeks understood perception. There is not a single straight line in the Parthenon: the floor bows upwards and the columns are fatter in the middle than at top and bottom. It is not designed to be perfect, it is designed to look perfect, counterbalancing the distortions of human vision. The vertical bars of the Rolls-Royce radiator grille use the same trick, entasis: fatter in the middle so they look parallel.

“It's not designed to be perfect, it's designed to look perfect.”

The point: Good design corrects for how people perceive, accepting objective imperfection to achieve perceived perfection.

For NHS Nightingale staff on 12-hour shifts, the team produced a chart for checking hydration by the colour of your urine. Rory says it worked on him personally for a year, though he warns that Berocca throws off the experiment badly, adding that in a power cut you could light your home with its eerie yellow glow. The serious point is about salience: in the normal course of things, the salient things are not important and the important things are not salient. Small interventions that make what matters salient at the moment it matters, as in factory safety work for Kimberly-Clark, are an overlooked design principle.

“The salient things aren't important and the important things aren't salient.”

The point: Good design surfaces what matters at the moment of action, because attention follows salience, not importance.

Rory asked a member what it cost to join the RAC Club. About £1,500 a year, he was told, which sounded incredibly expensive for a club. But compared with buying a flat in London it is roughly half the council tax, you can stay for £80 a night, and it has a swimming pool, a Turkish bath, three restaurants, two bars, a garden, a staff of 20 and its own post office. Nobody ever asks 'do I buy a flat or join a club?' because that comparison is too wide.

“Nobody ever looks at: do I buy a flat or do I join a club? That's just a little too wide to set the comparative net.”

The point: Price judgements depend entirely on the comparison set, and System 1 draws that set very narrowly.

Asked how an adman could persuade Donald Trump not to invade Greenland, Rory says you would have to do what someone once managed with a Roman emperor: persuade him that the real power play is not to do something. The emperor was told the most powerful thing you can do is not to have someone killed but to forgive them. You would have to reframe the whole thing in Trump's language.

“The most powerful thing you can do is not to actually have someone killed, it's to forgive them.”

The point: To change a powerful person's behaviour, reframe restraint as the stronger display of power.

Rory likes riffing with Nassim Taleb on things valued for reasons other than the ones people give. You buy a swimming pool so you can walk round your garden in a bathing costume without feeling odd; a dishwasher is mostly a cupboard for mildly unsanitary plates. Ask why people like the BA lounge and they cite whisky, sandwiches and Wi-Fi. But unconsciously the value is trust: you leave your laptop and bags on a chair and wander off for food or the loo. Moving from London to the country feels the same.

“The reason we buy dishwashers is to keep dirty plates out of sight.”

The point: Much of a product's value is unconscious trust and ease that customers never articulate.

Rory and his wife stayed at a very good-value Italian hotel with massively polarised reviews. The cause was the barman who brought toasted sandwiches, beer and ice cream to the pool. To British guests he was hysterically funny and charming, with a very dry, sarcastic sense of humour, an honorary Brit. Many American guests found him maddening, because being served with sarcasm is so incongruous with American service culture that they read it as rudeness.

The point: The same behaviour can delight or offend depending on the cultural frame the customer brings.

Rory mentions a viral video where a father hands his children an old dial telephone and asks them to call a number; they stare at it baffled for ten minutes. It struck him that we still use a dial phone with a receiver as signage pointing to a telephone, yet for anyone under about 15 or 20 it might as well be a Chinese character. He also worries that drivers who never drive a slightly bad car, like pilots over-reliant on autopilot, may lose instinctive skills.

The point: Symbols and skills depend on shared experience, and silently lose meaning when that experience disappears.

Straight after the story of the airport bus, which everyone assumes is rubbish simply because it is a bus, Rory offers his joke about how much emotional weight a few words carry. The three worst words in the English language are 'bus replacement service'. The three best words are 'all-day breakfast'.

“The three worst words in the English language: bus replacement service. The three best: all-day breakfast.”

The point: A few words of framing can carry an entire emotional verdict on an experience.

In an experiment Rory attributes to Daniel Pink's writing, two dogs share a box with an electrified floor that delivers occasional shocks. One dog has a button that stops the shock when it nuzzles it; the other does not, yet suffers exactly the same amount of pain. The first dog stays relatively content, while the second lapses into complete depression. The circumstances of our lives may matter less to our happiness than the sense of control we feel over them.

“The circumstances of our lives may actually matter less to our happiness than the sense of control we feel over our lives.”

The point: Perceived control, not just the objective outcome, drives wellbeing.

One of Rory's suggestions to the airport: although it is logically quite a good idea to have a lift with no up and down buttons if it only serves two floors, it is actually terrifying. When the door closes and there is nothing for you to do, you have stepped into a Hammer horror film.

“When the door closes and there's nothing for you to do, you've actually just stepped into a Hammer film.”

The point: A logically redundant control can be psychologically essential because it gives people a sense of agency.

We think things grow linearly and do not instinctively grasp compounding. Rory resented his early pension as holiday money doing nothing, until at 58 he wondered where all the money came from. A colleague of his met, in a pub, a man who had worked for him 20 years earlier, a deeply unemotional guy, and reminded him: 'Don't forget it was me who persuaded you to take out a pension.' The man hugged him. It had transformed his life and given him options he never knew he would have.

“We don't instinctively understand the compounding effect. We think that things grow linearly.”

The point: Selling something whose reward arrives decades later is a psychological challenge because we misjudge compounding.

Classical economics is preoccupied with reality, and reality is a poor guide to happiness. Pensioners and the young unemployed are in the same position: too much time, not much money. Yet pensioners are reportedly very happy and the unemployed very depressed, because pensioners feel they chose it. The English upper-middle classes solved this by rebranding unemployment as 'a year off'. A son unemployed in Manchester is embarrassing; a son unemployed in Thailand is quite an accomplishment.

“Having a son who's unemployed in Manchester is really quite embarrassing. But having a son who's unemployed in Thailand is really viewed as quite an accomplishment.”

The point: Our experiences depend less on what things are than on how we view them, especially whether they feel chosen.

A waiter offers two specials, fish or chicken, and the diner picks the fish. The waiter returns, apologising that he forgot to mention beef as well. The diner thinks and says, in that case, chicken. Rory explains that with only two options, fish looks like the healthy choice; adding beef makes chicken seem the virtuous middle option. A fast-food chain, he adds, could launch an enormous burger and everything else on the menu would suddenly feel like salad.

The point: Judgements are comparative, so adding an option can shift which of the existing choices looks sensible.

Responding to Rita McGrath's story of Zippo seeing its competition as men's gifts, Rory explains why men are hard to buy for: they are happy with luxuries, but only if the luxury has some notional function. There's no scented candle for men. The Swiss watch industry thrives on the same principle; nearly every complication can be done better by a phone, and if you want moon phases there's a free app, but the watch retains some mechanical function or utility that licenses the indulgence.

“Men are very happy with luxuries but they have to have some notional function.”

The point: A pretext of utility gives people permission to buy indulgences they couldn't otherwise justify.

Human behaviour doesn't run on objective reality; it is driven by meaning and context. The same thing in a different context can seem totally different and lead to totally different behaviour. Rory's favourite illustration is Mark Twain's Tom Sawyer, who gets his friends to pay him for the privilege of painting the fence simply by portraying fence painting as desirable.

“Human behaviour doesn't run on objective reality. It's driven by meaning and context.”

The point: Reframing a task can turn a chore into a privilege people will pay for.

Some travel websites deliberately slow their search, showing a screen saying they are searching easyJet, British Airways, Alitalia and others, then delivering results after 15 seconds of visible activity. People value those results more and book more readily than if they appeared instantly. Is it a con? Rory argues a business's job is to create perceived value: presentation affects how food tastes, and once food reaches table stakes, great restaurants win on atmosphere and theatre. The worst outcome, environmentally and commercially, is underappreciated value: making something great without letting people realise it.

“The worst thing you can do is to create underappreciated value.”

The point: Visible effort raises perceived value, and failing to communicate quality wastes the effort that made it.

A UK dog charity sent sponsors letters 'from' their adopted dog: thank you for adopting me, I'm gambolling in the garden. Some people adopted a second dog and got different letters, but in the same handwriting, and that they would not accept. The single best thing one person there did was create many different fonts so each dog wrote in its own hand. People happily accepted the implausible idea of a dog writing a letter, but two dogs sharing handwriting was a step too far.

“These people were totally happy with the idea that a dog was sending them a letter.”

The point: People accept obvious fictions willingly but reject small inconsistencies that break the illusion.

When Ogilvy was near Lambeth Bridge, a new employee asked which Tube station was closest and was told Embankment. For two years she travelled to Embankment, changing at Charing Cross every morning. Eventually a colleague asked why she did not just walk from Charing Cross. She had made the change some 400 times before realising the two stations are effectively next to each other.

“She goes, well, I change at Charing Cross every morning.”

The point: People follow the instructions and maps they are given and miss obvious shortcuts.

The self-checkout warning about an unidentified item in the bagging area is, for Rory, a classic silent irritant. Because it is spoken out loud, it is not just annoying but mildly humiliating: it broadcasts to everyone else at the checkout that this person does not know what he is doing.

“It's not just annoying. It's mildly humiliating.”

The point: Public error messages add social embarrassment to inconvenience, magnifying the irritation.

Rory credits Trump with something brilliant: because he speaks unscripted, off the cuff and slightly randomly, for an hour and a half without notes or autocue, watching him is like watching live sport rather than a recorded event; you feel anything could happen. Some of his tics, admitting he does not know things, nice person and nasty person, rudeness to journalists, belong almost to primatology, conveying an alpha-male sense of power.

“It makes it like watching a live sport rather than like watching a recorded event.”

The point: Unpredictability holds attention, and scripted communication sacrifices it.

When looking for a flat, use an isochronic map: put your workplace at the centre and slide the acceptable commute time. Places in south-west London inside the M25 show up as dead zones, while Harpenden on the Thameslink line, or Ashford in Kent thanks to High Speed 1, appears long before them. We assume distance is proportional to time, and nothing of the kind is true.

“Find out what the map's wrong about and you've got a competitive advantage.”

The point: Where the common model misleads, those who measure what actually matters find bargains.

Rory explains why consumers cannot tell you what they want. Citing Iain McGilchrist's divided brain and Michael Gazzaniga's work, he notes that language is largely a left-brain activity, so our capacity to accurately explain the origins of our emotions is very poor. An emotion is experienced in one part of the brain while another part confabulates a plausible-sounding story to explain it, a story that may have little to do with the real origin of the feeling or behaviour.

“The other part of the brain is effectively confabulating a plausible-sounding story to explain the emotional state.”

The point: Stated reasons are often post-hoc stories, so research must look beyond what people say.

Rory argues our problem is not a lack of marvellous things but a lack of appreciation for them. He invokes a classic definition of poetry: making new things familiar and familiar things new. Advertising can do the same, getting people to look at existing things afresh instead of wanting more of something else. On the complementary-goods view, its job is to add appreciation to what already exists, extracting value rather than adding it.

The point: Raising appreciation of what people already have can create value without making anything new.

Rory, a small investor in Wispr Flow, thinks mass voice dictation would hugely raise productivity, but people don't realise how much faster it is until they try it. We think of ourselves as fast typists and never reframe widely enough to ask what else could be much faster. He suspects BlackBerry and Steve Ballmer were right that work devices need keyboards, but when the lovely iPhone arrived we decided we were quite fast at screen typing because we had nothing to compare it with. AI dictation even learns context, getting 'Llandovery' right.

“What we're not doing is reframing widely enough to go: what else could I be doing that's a hell of a lot faster than typing?”

The point: Without a frame of comparison, people accept a slow method as fast.

Rory solved mask-wearing for himself: if you wear a hat, wearing a mask with it makes you feel like an international man of mystery, which completely changes the emotional association of being masked. He also thought masks should have had elastic that went round the neck, so you could wear them like a cravat when not covering your face.

“Wearing a mask with a hat makes you feel like an international man of mystery.”

The point: Changing the emotional association of a behaviour can make a chore feel desirable.

The Manifest advertising festival, free and by invitation, had hard data showing a steady uptake of about 15 per cent of invitees. Then one year there was mayhem, with queues around the block and take-up of around 40 per cent. It had crossed the threshold of being a thing: from "I wonder if I'd go to Manifest" to "Are you going to Manifest this year?", from a default no to a default yes. Food diffuses similarly slowly: a 1960s friend said the two important questions were Indian or Chinese, and Beatles or Stones.

“It goes from a default no to a default yes because it's just become a thing.”

The point: Adoption is non-linear: once something becomes a social default, uptake jumps.

In the Meredith Kercher case, investigators fixed early on the idea that a break-in had been staged, because entering via an exposed upstairs window seemed absurd. They privileged that hypothesis and ignored contradictory evidence. But the point of smashing the window was not access: it was to make a lot of noise from a spot where escape was easy. If nobody reacts, you can safely climb in five minutes later; if a light goes on and a dog barks, you leg it. Like the mouth, which also lets you breathe when your nose is blocked, things in complex systems serve several purposes.

“The purpose of breaking a window was not to gain access to the house, but to make a hell of a lot of noise.”

The point: Behaviour that looks illogical often serves a hidden purpose, and fixating on one theory blinds you to it.

In the style of Robert Frank's economics puzzles, Rory asks why there are more fish restaurants by the sea. The business answer is supply: fresh fish nearby. But then there would be plenty five miles inland, and the cheapest fresh fish is probably at Billingsgate in London. His explanation is psychological: fish tastes better by the sea. And a complexity one, quoted by Nassim Taleb: in any party of six someone does not fancy fish, so consensus is hard, except at the seaside, where everyone wants fish at once. The supply story is 'rigorous but wrong'.

“Fish tastes a lot better when you're by the sea.”

The point: Neat economic explanations can satisfy our need to look rational while missing the real, psychological why.

Rory's mother never went to university but was an untouchable amateur psychologist who could spot an affair at 500 paces; Rory's own antennae are weaker and he has had to learn. People slightly on the spectrum make good social scientists because they work things out from first principles. A neurotypical person just knows you make an effort for a wedding. A purely rational person asks why not wear an anorak: the forecast says rain and it has lots of pockets for nerdy things.

“Why is it inappropriate to wear an anorak to a wedding?”

The point: Questioning norms from first principles reveals the hidden social logic others take for granted.

Rory thinks KFC superior to McDonald's because McDonald's is prescriptive: this is what you eat and this is how you eat it. At KFC you can mix the gravy with the fries and everything else. At a far more elevated level, haleem comes with little bowls of chilli, ground ginger and other toppings so you can vary the flavour as you go. Personalisation, the host notes, is half of Nando's genius too. Rory adds that variable reinforcement makes the experience more interesting.

“You get variable reinforcement, which is more interesting.”

The point: Letting people customise and vary what they consume raises engagement and perceived value.

Rory's mischievous theory: advertising people, supposedly dissolute, are mostly devoted family types. For a politician or a US televangelist the risk makes transgression five times more enjoyable; an ad man caught in a motel with a girl called Tammy is mildly embarrassed. A friend from the music industry was never so shocked as at his brother-in-law's provincial accountancy Christmas party, their one chance a year to let loose. Early punk gigs were more violent in the provinces, where everyone felt they had to outdo London.

“The element of risk makes transgression five times more enjoyable.”

The point: Scarcity and risk raise the appeal of transgression, so the most constrained may misbehave hardest.

When Les Binet says that if he could eat only one cheese for ever it would be cheddar, Rory calls that an honesty heuristic: cheddar is the most popular cheese, and he distrusts anyone who despises it in favour of something French, which suggests a cheese snob. His own joke, which he believes later appeared in an advert: the reason the French have 150 different kinds of cheese is that they still have not come up with one as good as cheddar.

“The reason the French have 150 different kinds of cheese is because they still haven't come up with one that's as good as cheddar.”

The point: Openly liking the popular choice is a sign of honest taste rather than status-seeking.

Most data does not capture lived experience over time, which Rory finds epistemologically dodgy. At a casino, if you have a long losing streak and then win big on the last hand, you leave punching the air. If you win massively on the first hand and lose gradually all evening, you leave feeling like an idiot. The order in which things happen plainly affects mood.

“The order in which things happen patently has an effect on human mood.”

The point: Sequence shapes how experiences are felt and remembered, which snapshot data ignores.

Rory says his electronic cigarette has given him untold happiness, and not mainly because of the nicotine. Since the UK smoking ban he had stopped enjoying drinks parties. Talking endlessly is tiring, and sometimes you just want to stand alone and stare out of the window. Without a cigarette, doing that makes you an antisocial, friendless idiot. With a cigarette, you are a philosopher. Same activity, small change of posture, opposite feeling.

“If you stand and stare out of the window on your own with a cigarette, you're a philosopher.”

The point: The power of reframing cannot be overstated: identical behaviour feels great or terrible depending on its frame.

Rory's theory: take away social embarrassment and all Englishmen would wear shorts all the time. On holiday among strangers, where it's normal, they wear shorts every day, because shorts are simply more comfortable. He reckons 60 to 70 percent of people would be naturists: alone in an unoverlooked villa with a pool in warm weather, would you put on trunks? Of course not. A huge amount of human behaviour is constrained only by how it will look.

“If you take away social embarrassment, all Englishmen would wear shorts all the time.”

The point: Much behaviour reflects social sanction rather than preference, which hides latent demand.

Some things become completely different out of doors. An Indian takeaway eaten as a picnic is weird and wonderful. If you put your router in the window and work in the garden in summer, it does not really feel like work. Rory notes that BT will send you a free mesh disc if you say you have weak coverage in part of the house. You can create a complete contextual shift just by moving outside.

“If you go and work in the garden in the summer it doesn't really feel like work.”

The point: A change of setting can transform how an activity is experienced.

An Ogilvy client said it is perfectly possible to manufacture a fly spray that smells of lavender, a fly spray and an air freshener in one. You can make it, but you can't sell it. The brain wants a trade-off: if it smells pretty nasty to me, it must be a real bastard to a fly. Rory is likewise suspicious of Nurofen Meltlets, which taste too nice to be effective; analgesics should taste horrible.

“It smells pretty nasty to me, so it must be a real bastard to a fly.”

The point: Consumers read unpleasantness as proof of potency, so removing the downside can remove the credibility.

It is very difficult to evangelise things that happen slowly or gradually without sounding late to the party. Rory thinks video conferencing is hugely important, and growing up on early Star Trek, someone appearing on a screen to talk to you was science fiction. But because it has been kicking around in unsatisfactory form since Skype, around 2001 or even 1999, and improved and was adopted gradually, saying 'isn't video conferencing brilliant?' sounds absurd. So we downplay its importance.

“It's very difficult to evangelise things that happen slowly or gradually without sounding late to the party.”

The point: Gradual innovations are systematically undervalued because there is never a moment when praising them feels new.

Ask Mr Logic how to compete with Coke and he says: a drink that tastes nicer, costs less and comes in a bigger can. Nobody gets fired for that. Yet the most successful challenger to Coke in 150 years comes in a tiny can, costs a fortune and tastes disgusting. A research firm that only tests carbonated drinks called it the worst drink in its history; one tester said he would not drink it if paid. But flip the context from drink to drug and those three qualities make a brilliant placebo. Medicinal products have to taste weird, the small can says the stuff is potent, and the price says it works. It was marketed as a drug: it gives you wings.

“If you flip the context from drink to drug, those three qualities which make Red Bull an appalling drink make it a brilliant placebo.”

The point: Qualities that are flaws in one category can be credible signals in another, so making a product nicer or cheaper can undermine belief in it.

Variants: Rory also uses it to show no corporation would have invented Red Bull: it would never survive a PowerPoint or a business case. Some tellings pair it with Starbucks as things nobody asked for, or with Huel, which should taste fine but not delicious.

When we marry, we make our vows in church in front of everybody we know at once, not door to door or by email, because it only takes one person to answer 'just cause or impediment' and everyone knows you are a liar. A promise made in public carries more conviction than one made in private. And we do something really expensive up front: an engagement ring has to hurt the man buying it, as proof of long-term intention, because nobody buys one for a one-night stand. Even the invitation could informationally be an email, but nobody sends one that way. Advertising works the same way: you only advertise expensively and publicly if you expect repeat customers, which is why ever more targeted, efficient advertising strips out trust.

“You wouldn't buy an engagement ring if you're only planning a one-night stand.”

The point: Messages are believed in proportion to what they cost the sender and how publicly they are made, so cheap, private communication earns little trust.

Variants: The cost need not be money: a musician who sings the invitation, rhyming the church postcode, has paid in talent. Some tellings credit Don Marti with the public-vows point; one frames it as a law of commensurability, where a big decision deserves a significance marker, so a car buyer wants a brochure. Rory once joked in India that 8 per cent GDP growth would all be spent on weddings.

Rory's American Express card reads 'Member since 95'. Functionally it does nothing; it just says we know how long you have been a customer. Probably suggested by someone at Ogilvy in the mid-1960s, it turned a transaction into relational capitalism. Nobody wants to cancel, because rejoining would reset the date to the current year and make them look like a Johnny-come-lately, so it exploits sunk cost too. When Amex mistakenly sends long-standing members replacement cards showing the current year, a large share ring up demanding their old date back, though nothing material differs. People at Amex told Rory it has been worth billions. He tried to find out what Ogilvy was paid for it: somewhere between nothing and $50.

“There are certain things where you can make people feel great which cost nothing at all.”

The point: Recognising how long someone has been with you costs almost nothing and builds loyalty that transactional accounting cannot see.

Variants: The share who complain is put at around 40 per cent, a high double-digit percentage, or more than half. Rory pairs it with boarding first (someone has to, so make it a privilege), BA lifetime tier points, and the bonus that a 19-year-old with a 'member since 92' card is probably holding a stolen one. He also uses it to argue marketing is fat-tailed.

Rory always wanted someone to take a superb product and market it atrociously. Two Melbourne comedians did: they put Ed Sheeran, the hottest musical act in the world, fully clothed on a stool inside a seedy peep-show booth, with a shady tout outside offering passers-by 30 seconds for two dollars. People assumed it was a code word for drugs or something worse. It took two hours and 23 minutes to find anyone brave enough to go in. The tout's first instinct was the default economist's suggestion: maybe we're pricing it too high. What finally worked were behavioural tricks: he's only here till midday, it'll get busy later, you can both come in. One ad in a respectable newspaper could have sold it at $200 with a queue round the block.

“If you have a great product but nobody trusts you, you don't have a great product.”

The point: Context and trust, not price or quality, decide whether a product sells, so bad presentation can destroy as much value as good presentation creates.

Variants: Rory often aims it at financial services: that is how it feels to market a good product nobody trusts. He also uses the tout's price-cut instinct to warn that algorithms will do the same.

Rory believes he coined it, having googled it: a flower is a weed with an advertising budget. A bee can only find out whether a plant has worthwhile nectar by visiting it, so plants need a reliable signal, and big petals are one. Because petals are an upfront cost, which also makes the plant likelier to be eaten, they indicate the seller's confidence: a flower that did not expect the bee to come back would not bother to grow them. Cheat the bee and it won't return. Some flowers also produce scents that only plants with plenty of pollen can make, and bumblebees have evolved to prefer them. If creatures with a bee's IQ can work out signals that are reliable because they are costly, humans surely have too.

“A flower is simply a weed with an advertising budget.”

The point: The cost of a signal is itself the message: extravagance is credible because only a seller expecting repeat business can afford it.

Variants: Sometimes paired with the line: efficient advertising can find customers, effective advertising creates them.

Rory asks a room of middle-class Londoners how many have been to Machu Picchu: 10 or 15 hands go up, or everyone if they are Gen Z after a gap year. Then how many have been to Lincoln Cathedral: two, usually people who were born or studied there. Yet Lincoln is a day trip by train for about £23, needs no long-haul flight and no altitude sickness, and architecturally it beats Machu Picchu. From 1311 its spire made it the tallest building in the world, overtaking the Great Pyramid for about two centuries, and the castle next door holds an original Magna Carta. It just lacks bragging rights. Much tourism is going somewhere far away to show off, and difficulty of access has become the status signal.

“An awful lot of tourism is people going somewhere far away for bragging rights.”

The point: When a choice doubles as a status signal, difficulty and distance get valued over intrinsic quality, and nearby wonders are overlooked.

Variants: The ratio varies by telling, from two or three to one up to about seven to one. Rory adds that there is no correlation between how elaborate a holiday is and how much you enjoy it, and that for the young, travel has replaced the car as the status symbol.