Rory Sutherland and Tom Goodwin Fix Marketing (Part 1)

creator | channel

Abstract

The first half of a long three-way conversation between a host and two marketing thinkers, opening on the claim that the buyer's problem is abundance rather than scarcity — more choice damages confidence in the decision rather than improving it.

• Preference is created by small details discovered in use, not by the specification on the box.
• A range nobody can build a mental matrix of sells nothing at all.
• The most valuable work sits outside the advertisement, and clients cannot buy it because no job title owns it.
• The performance playbook fitted about a fifth of clients and spread because it was new and defensible rather than because it works.
• Marketing is fat-tailed, so averaging destroys exactly what you were looking for.
• Surprise multiplies whatever feeling the work already carries, and it is the one effect nobody has to buy.

For a person building a personal brand, the hour reads as a catalogue of constraints that mostly do not apply. Every pathology described here — the veto, the demand for advance justification, the quarterly horizon, the averaging that deletes the outlier — belongs to an organisation with separate functions. A solo operator holds all of them at once, which is why the arbitrary gesture, the long-dated bet and the one surprising element survive intact rather than being reasoned out of the work.

Chapter summaries

00:53 - Abundance rather than scarcity is the buyer's problem, because more choice damages confidence in the decision

Abundance rather than scarcity is the buyer's problem, because more choice damages confidence in the decision.

"to be alive today is to be overwhelmed. And I think we live in the age of abundance, where we haven't realized that more choice is not a good"

— 00:53

People choose to minimise the risk of later regret, and profusion makes that confidence impossible to reach. Retailers already practise reduction by burying the second page of results; shoppers practise it by preferring places with limited stock. The paradox holds where nobody prepared for it — someone who travelled to a specialist shop is not overwhelmed by its range.

02:32 - A unique selling proposition works by manufacturing importance and granting permission to buy

A unique selling proposition works by manufacturing importance and granting permission to buy.

"We focus on the one thing that our product has that nobody else's product has. We amplify that thing to"

— 02:32

Amplifying one attribute makes it important simply by being attended to, and importance then licenses the purchase. The proposition is not really a claim about the product: it is a permission slip, and the mechanism is attention rather than evidence.

03:52 - Preference is created by small details discovered in use rather than by the specification

Preference is created by small details discovered in use rather than by the specification on the box.

"there's something there that can inform new product development in the sense that adding what you might call Easter eggs to a product"

— 03:52

An unmentioned feature encountered in daily use produces loyalty that no listed attribute earns. A button that revives cold toast; a seat that inflates through corners. Neither was ever advertised, and both decided a repeat purchase — which is why building them in is product development rather than marketing.

05:30 - The unique selling proposition was never crude, because it works by dramatising what buying elsewhere forfeits

The unique selling proposition was never crude, because it works by dramatising what buying elsewhere forfeits.

"the sort of example of a little thing that made a big difference that no advertising campaign ever talked about to my knowledge"

— 05:30

Framing the claim as a loss makes it bite harder than the same claim framed as a gain. What the proposition really dramatises is what you forfeit by buying elsewhere — which is why disparaging it as crude turns out to have been premature.

06:58 - Companies destroy their own choice architecture by proliferating brands customers cannot tell apart

Companies destroy their own choice architecture by proliferating brands customers cannot tell apart.

"there are some cases where unwittingly someone has just destroyed the choice architecture"

— 06:58

A portfolio whose members are indistinguishable leaves the customer unable to place any of them. Dozens of sub-brands nobody can rank is not a range but a fog, and loyalty programmes end up being used mainly to make the decision smaller.

12:37 - Consumers run one simple model: dearer is better, and occasion picks the spot

Consumers run one simple model — you get what you pay for — and pick their spot on it by occasion.

"of the time the consumer logic is you get what you pay for. There's more expensive and better, and there's cheaper and maybe a bit crap, and you want to find your space."

— 12:54

The line runs from dearer and better to cheaper and probably worse, and where you sit is decided by how much this category matters on this occasion. Position is chosen by occasion rather than by wealth — a format right for two nights is wrong for seven and wrong again for a honeymoon.

14:40 - The most valuable work sits outside the advertisement, and no job title owns it

The most valuable work sits outside the advertisement, and clients cannot buy it because no job title owns it.

"The thing that really annoys me about advertising agencies is that they they they confine their creativity to such a narrow range of activities"

— 14:40

Nothing gets done without a role attached and a purchase order raised, so work that falls between departments is unbuyable however valuable it is. Ordering screens that could be worth fortunes are never experimented with, and nobody is the director of the small annoying things.

18:26 - Numeracy is scarce enough in marketing that ordinary school mathematics becomes a large advantage

Numeracy is scarce enough in marketing that ordinary school mathematics becomes a large advantage.

"the number of times where we will talk about campaigns that we've seen and I always wonder whether people in the real world have done that"

— 18:26

Deliberate exposure to ordinary life builds a corpus against which nonsense in a meeting becomes audible. The suspicion is straightforward: campaigns discussed at industry events may never have been encountered by anyone outside the room.

22:00 - Original thinkers are not thinking differently, they are experiencing differently

Original thinkers are not thinking differently, they are experiencing differently.

"I don't think I think differently. I think I I think I experience things differently."

— 22:00

Thinking differently is the wrong description. Buses, cheap hotels, discount pharmacies and shopping centres are the method — and what then reads as contrarianism is only someone asking whether the thing just said is true. Experience is the input, not temperament.

23:52 - The performance playbook fitted a fifth of clients and spread for unrelated reasons

The performance playbook fitted about a fifth of clients and spread on properties unrelated to whether it works.

"wastage is an amazing drug cuz it basically allows you to reach people who in 10 years time or after a promotion or when they get old will buy your stuff"

— 23:52

It was new, sophisticated-seeming, data-rich and defensible, and those are the properties that made it spread — none of them being whether it works. Reaching people who will buy in ten years is a feature rather than waste, and a playbook built for small companies with short horizons was copied by large ones with long ones.

26:50 - Marketing is fat-tailed, so averaging destroys exactly what matters

Marketing is fat-tailed, so averaging destroys exactly what matters.

"That when you aggregate information and you average it, you lose what's really significant and interesting in the act of averaging"

— 26:58

A tenth of the work delivers most of the value and cannot be identified in advance, so aggregation removes precisely the outlier you were looking for. An accidental element wandering into a photograph became a brand property worth a fortune. Averaging is the enemy of anything fat-tailed.

29:20 - The instructions that actually produce great work cannot be entered into a corporate process

The instructions that actually produce great work cannot be entered into a corporate process.

"One advantage of being 60 is you're less worried about looking like an idiot"

— 29:20

Be lucky, be patient, trust your judgement — all true, all useless inside a process built to ratify decisions. The habit of codifying a successful accident into ingredients and replaying them causally produces the ingredients without the accident.

31:03 - The devices that build memory are already evidenced, and the industry is simply not using them

The devices that build memory are already evidenced, and the industry is simply not using them.

"make the point that animals are more likely to capture attention, create an emotion, and create a long-lasting memory than anything else"

— 31:11

Decades of data show which devices capture attention, create emotion and build lasting memory. They remain unfashionable anyway — which is why nobody in the conversation can name a tagline introduced recently, while the old ones are still there.

32:14 - Marketers answer for every penny of cost and keep almost none of the upside

Marketers answer for every penny of cost and keep almost none of the upside.

"marketers are ill-treated by the finance function because"

— 32:14

The breakthrough gets credited elsewhere in the annual report and the fee for it was a rounding error. Held responsible for every penny of cost and allowed a fraction of the upside for one quarter — described plainly as not a reasonable way to assess any part of a business.

33:55 - The industry has no working memory, because today's sales are decades-old advertising

The industry has no working memory, because today's sales are the residue of advertising bought decades ago.

"the brands that I now buy, the center of gravity of the ads that I will have seen um will go back to media buys"

— 33:55

The people who bought the advertising that is still working are gone, so nobody in the building is connected to the thing currently producing sales. Present-day buying habits trace back to media bought decades ago, and the only person who remembers the campaign that changed a behaviour is the one who was there.

36:37 - When five per cent of successes pay for everything, a short run carries no information

When five per cent of successes pay for everything else, a short run of results carries no information.

"that's true of all fat-tailed businesses. So, archetypal fat-tailed businesses would be venture capital"

— 36:37

In a fat-tailed business three bad outcomes are noise, and dismissing someone on them is responding to nothing at all. Successors routinely profit from decisions their predecessor made before being fired for the results of decisions someone else made. The signal is not in the run.

37:51 - Forcing a return on investment onto every decision forecloses luck

Forcing a return on investment onto every decision forecloses luck.

"if you're famous, people come to you with things. In other words, if you're famous, customers find you."

— 37:51

Being known means unknown people bring things to you rather than your having to find them, which is a phase transition rather than an improvement. And the choices that produce it pay only probabilistically, so demanding a return on each one forecloses the whole mechanism.

39:52 - Who owns a company predicts whether its advertising works

Who owns a company predicts whether its advertising works, and the listed form may be incapable of marketing and innovation.

"You know, they realize that these things really, really take time."

— 40:18

A family-owned business can run several time horizons at once; a quarterly cycle optimises for one variable and calls it success. Four of five recent effectiveness award winners were family-owned — and nobody would accept a single metric over a single period as a measure of their own life.

45:03 - What a customer feels is the gap between the promise and the arrival

What a customer feels is the gap between the promise and the arrival.

"I was fascinated I had this conversation near I hour recently on the podcast talking about how our prediction of something influences how we feel"

— 45:03

Reward is registered against prediction rather than against the thing itself, so what a customer feels is the gap between the promise and the arrival. Dopamine arrives at the signal rather than the reward; two airlines with inverted complaint rates have inverted promises rather than inverted service.

48:15 - Surprise multiplies whatever feeling the work already carries, and nobody has to buy it

Surprise multiplies whatever feeling the work already carries, and nobody has to buy it.

"database. They looked at every single reason everyone's ever given in a survey why they're surprised."

— 48:27

An unexpected act costs little and scales to every level of an organisation, which makes it the one marketing effect nobody has to buy. What people are surprised by is almost always the product against expectations — and the two occasions anyone returned a missed call are remembered years later. Surprise is the multiplier.

51:33 - Making effective advertising may be straightforward, but its surprising part cannot be reasoned out

Making effective advertising may be quite straightforward, but its one surprising part cannot be reasoned into existence.

"maybe if you're making ads, like maybe you get the thing that you're making and you explain explain it in a nice way"

— 51:50

Explain the thing well, add a line, add music, ask for the action — and then add something arbitrary. The basic playbook may be most of it, and the part that cannot be reached that way is the surprising part, because it is difficult to be rationally surprising.

53:21 - Commercial innovation leaves the product alone and changes how it is sold

Commercial innovation leaves the product alone and changes how it is sold.

"innovation where the product itself is not changed, but the way in which you sell it, present it, price it, charge for it"

— 53:37

Presentation, pricing and packaging can be changed where the product cannot, and the returns are comparable. Names on packaging, payment in instalments, ordinary goods sold in sizes that leave nothing behind — none of it touches the thing being sold.

54:59 - Heavy process reaches the obvious answer expensively and strips out the best work

Heavy process reaches the obvious answer expensively and strips out the work that would have won biggest.

"all these organizations, they're so defensive in their decision making"

— 54:59

A long rational selection arrives expensively at what was obvious at the start, and removes anything that could not survive the deck. Nobody could choose a partner or a home that way, and no organisation would produce anything interesting if the justification had to come first.

56:41 - Complexity is a commercial artefact, and design fails at the decisive moment the infrequent customer returns

Complexity is a commercial artefact, and design fails at the decisive moment the infrequent customer returns.

"Actually quite a lot of government websites are actually quite good, aren't they?"

— 56:41

Services built to extract rather than to serve become slightly better on average and unusable exactly when they are needed. An application that loses a login during an update fails at the one moment a year the customer opens it — while the ones with nothing to extract simply ask what has to be done.

58:49 - Hourly billing pays for time, and the cost-justification layer it grew now outweighs the work

Hourly billing pays for time, and the cost-justification layer it grew now outweighs the work.

"the worst thing that happened to marketing services was payment by the hour"

— 58:49

Time consumed and value added are barely connected, so the administration built to prove the cost grows larger than the work. A senior lawyer spends three times as long justifying charges as doing the job, and the headcount that falls is the billable kind rather than the unbillable. The layer outweighs the thing.

Personal Branding Lessons

Almost every constraint in the hour above is organisational: the purchase order nobody will raise, the finance model that averages the outlier away, the process built to make a choice defensible before it is made. A solo operator has none of that machinery and plays all of its parts anyway — you are the only one who can veto your own idea, the only one who will ever ask you for proof, and the one asset in the business that cannot be bought, copied or reassigned. Here is what the episode hands you to use.

Name the thing a reader gives up by following someone else instead of you. Amplifying one attribute does two jobs at once: attention makes that attribute the axis the decision gets settled on, and declaring it closes the comparison so nobody has to keep looking. The under-noticed half is loss aversion — the same claim stated as a forfeit bites harder than the same claim offered as a gain, which is why the thing only you can truthfully claim should be written as a sentence, not implied across a feed. You will know it worked when a stranger introduces you to someone using your sentence, not their own. 05:30

Build in something you never mention anywhere. The toaster button and the seat that inflated on corners were never the subject of a campaign, and that is the mechanism: a detail the owner discovers arrives as a gift, because no expectation was set for it to live up to. Yours costs nothing to install — a real reply to every first-time subscriber, a working tool buried at the foot of a long post, an answer to the question everyone else ignores. The outcome to picture is someone describing a thing they found, in their own words, without your marketing attached to it. 03:52

Price on a single line and state what the dearer option actually buys. Nobody holds the two-by-two matrix you would build; the working consumer logic is more expensive and better, cheaper and a bit crap, find your spot. Two products at the same price with nothing encoded between them sold nothing at all, because confidence collapsed before the decision did — which is the whole of what people are doing when they read your prices. Get it right and the email asking what the difference is stops arriving; what comes instead is people naming the tier they want. 12:37

Make yourself director of trivia this week. Inside a company the small annoying things fall between too many stools — no job title owns them, and procurement will not raise a purchase order for the person who fixes them. You have neither obstacle and no excuse: the bio that never says what you do, the sign-up that sends nothing for six days, the link in the pinned post that 404s. The consequence is specific and countable — the reader who reached your sign-up form and left does not leave. 14:40

Spend an afternoon a week where nobody in your niche has any reason to be. What reads as an unusual mind is often just an unusual set of inputs: buses, bad hotels, the strange pharmacy, the shopping centre — raw material that someone travelling by car and staying in good hotels never receives. Go and look, then write down what you saw before deciding what it means; documenting it as it happens is cheaper than trying to recall it later. What you get is a post none of the forty accounts in your category could have written, because none of them were standing there. 22:00

Judge your output over forty attempts, never over three. Roughly a tenth of creative work delivers the overwhelming share of the value, and which tenth is not known in advance — which is why sacking a manager after three defeats is a judgement made on noise. Decide your number of attempts before you publish the first one, and treat the flat ones as the cost of staying in a game where the payoff is lumpy. The consequence worth picturing is being still there in month nine, when the one that travels finally goes. 36:37

Pick one thing about your work by taste alone and refuse to defend it. It is difficult to be rationally surprising: what is left over after the reasoning is arbitrary and subjective, the two qualities a justification process exists to remove. Any choice you can prove on paper is a choice your competitor's reasoning also reaches, which is the quiet argument behind every audit of what actually separates you. No committee will ask you to compare the teddy bear against the stuffed aardvark, so install the gesture by fiat — the odd sign-off, the format nobody else uses — and it becomes the detail people use to describe you when you are not in the room. 51:33

Set the expectation below the thing you are actually going to do. Feeling is a prediction error: complaint measures the gap between the promise and the arrival, not the quality of the arrival, which is why the legacy carrier is complained about constantly and the budget one hardly at all. Say the reply takes three days and answer on the first; tell people when the thing will land rather than leaving them to guess. What arrives back are thank-you notes for work you were always going to do. 45:03

Go back to the one that got away. Surprise multiplies whatever emotion the work already carries, and it costs almost nothing — the agent who rang the hotel three days later unprompted and got a night refunded, the salesman who noticed a missed call from the forecourt and phoned back, twice in twenty years of caller ID existing. Your version is the unanswered message from three weeks ago, the person you could not help, the refund nobody asked you for. That is how the handful of people who tell everyone about you get made: a screenshot of your reply, posted by someone you never asked for anything. 48:15

Questions

Each answer ends at the moment in the recording where it is given.

Each answer ends at the moment in the recording where it is given.

Why do people want fewer options, not more?

Confidence is the casualty, not effort. Each extra option adds another thing the buyer could later wish they had picked instead, so the risk of regret climbs while the quality of the eventual pick barely moves; eighteen of something is a tolerable amount of thinking, seventy-five is paralysis. What people are minimising is the not-knowing, which is why narrowing the range only helps when the remaining options are explained — an unexplained shortlist buys nothing. 02:17

Why do the best known brand assets arrive by accident?

Nobody can commission a detail whose value is only visible once it exists. The sheepdog that became a paint brand's signature walked into a shoot uninvited, and the only decision that mattered was someone on set judging the picture better with it in. The mistake comes afterwards: codification, treating the dog as the ingredient, when what is reproducible is the noticing — which is why working in the open, where accidents can be seen and kept, beats briefing for them. 29:55

Why does averaging destroy what you are looking for?

Because the mean is computed by deleting the extremes, and in this business the extremes are the entire return. A small share of creative work produces most of the value, and which share is not known in advance, so an aggregate reports the typical case — precisely the case nobody is trying to produce. The limit: averages stay useful for efficiency work, which keeps the lights on; they simply cannot say whether the rare outsized result is any nearer. 26:54

Why does expectation decide satisfaction?

Feeling tracks the gap between what was predicted and what arrived, not the absolute standard of what arrived. Attention is spent on deviation from the brain's own forecast, which is why the budget airline is rarely complained about and the legacy flag carrier constantly is. The trap is inflating anticipation to win the sale: what is charged and claimed becomes the standard the delivery is then marked against, and a promise kept modestly beats a promise oversold. 46:31

Why does the payoff arrive after the person who paid for it has gone?

Because what the money buys is a memory, and memories are cashed in years later by people who have forgotten where they came from. A supermarket basket today reflects impressions bought in the 1980s and 1990s; whoever authorised them has long since moved on. The asymmetry is what makes short spending rational for the individual and ruinous for the firm — returns that compound land on a successor, while every penny of cost is charged to the incumbent. 33:48

How do I find my counterintuitive insight?

By changing the inputs, not the reasoning. Unusual conclusions mostly come from unusual exposure — buses, cheap hotels, the strange pharmacy, the ordinary places that a car and a good hotel quietly remove from a life — and that accumulated first-hand material is what a meeting's claims can be tested against. The under-rated half is nerve: noticing without asking the plain question changes nothing, and the ground nobody else is standing on is only worth anything once someone names it aloud. 22:00

Why does the ending decide the memory?

Recall keeps a few snapshots rather than a running total, and the final slot is always filled. Competence through the middle largely disappears, while a deviation at the close — a refund chased down three days after the interaction had already failed — becomes the whole account of what happened. The condition is surprise: a close that matches the forecast is compressed to nothing, so the moment that captures attention has to break a prediction, not merely meet one. 50:20

Why is being famous the whole game?

It reverses which way the effort runs. An obscure operator spends their life finding customers; a known one is found by them, and the switch is probabilistic rather than causal — recognition enlarges the surface area on which good things can land. Worth holding the limit: what is gained is optionality, not certainty. Fame multiplies the number of approaches, and standing that a stranger can act on still has to be there when one arrives. 37:49

Sources

Rory Sutherland and Tom Goodwin Fix Marketing (Part 1)

The sections above follow the episode's own order. Each timestamp is the point in the recording where the idea is discussed, so the post reads as a map of the conversation rather than a rearrangement of it.

Quotations come from the episode transcript with spoken filler ("okay", "right", "you know", "um", "actually", "kind of", "sort of") and false starts removed. No words are added: square brackets mark anything inserted for sense, and an ellipsis marks any cut. British spellings are restored, and speech-to-text errors were repaired against context only where the intended word is unambiguous. Where a line appears both in a cold-open teaser and in its place in the conversation, the timestamp cites the second — the point at which it is actually said.

Quotes are stamped with the moment they are said rather than with a speaker's name.