· Rory Sutherland

Killer marketing secrets that always work (ft. Ogilvy Adman, Rory Sutherland)

There are two ways to make money — make desirable things or make things desirable — and because the laws of psychology are far more malleable than the laws of physics, reframing perception is often cheaper and higher-leverage than improving the product.

marketingbehavioral-economicsbrandingpsychologypositioningpricing0% confidence

Why this is in the corpus

Rory Sutherland (Vice Chairman, Ogilvy; author of Alchemy) delivers a dense masterclass in behavioral marketing: reverse-benchmarking, explicit trade-offs, value capture via outcome pricing, psychological bottlenecks, and the fat-tailed nature of innovation — a goldmine of non-obvious, generalisable operator doctrine.

Summary for skimmers

Two ways to make money: make desirable things or make things desirable. Reduce anxiety instead of increasing range. Reverse-benchmark the neglected metric. Make trade-offs explicit and customers accept them. Marketing and innovation are fat-tailed. Direct response was behavioral economics before the name existed.

Briefing

What survives the editorial filter

This page should feel like a smart colleague already listened for you and left only the operating logic worth keeping. Not everything said in the episode makes it through.

Trust signal

Direct episode extraction

Best used for

Decision-grade retrieval metadata not yet added for this episode.

Hold lightly

No explicit downgrade reason stored yet for this episode.

Principles

Durable claims that survive beyond the speaker's biography — each with explicit limits, transferability judgment, and evidence.

Principle

Reason should be the slave of the passions — deploy logic to sense-check, not to originate

Let intuition originate and reason verify, not the reverse.

Like a detective who may only use evidence and never a hunch, a purely logical firm can reach only ~5 of 25 possible solutions — all already occupied by rivals using the same stepping stones.

Use data to correct emotional reasoning when it errs, not to forbid intuition.

Principle

You can't market explicitly to the old, so much valuable tech never reaches them

Aspiration-driven imagery orphans older buyers from products built for them.

The average new VW Golf buyer is ~59 but ads show 28-year-olds; bone-conduction headphones and folding phones are godsends to the old yet marketed only to joggers and youth.

If your real buyer is old, reach them without labelling them old.

Principle

Focused targeting surfaces unexpected incremental audiences

Narrow focus produces distinctiveness that draws in audiences you never designed for.

Moxy (Marriott-on-Gen-X) drew 60-year-old couples for short stays; Flat White or Off, aimed at stations, found its best market in conferences.

Commit hard to one audience and watch for the surprise segments distinctiveness attracts.

Principle

Give your secrets away — you'll be copied far less than you expect

Publishing your playbook wins authority because rivals are culturally unable to copy it.

Ogilvy published How We Write Ads; direct mail evidence rarely gets copied because competitors find old-fashioned media too unfashionable to justify.

Teach how you think publicly — the moat is cultural incapacity, not secrecy.

Principle

Drawing attention to an attribute makes customers reweight their utility function

Make a neglected attribute salient and it becomes important to the buyer.

Buc-ee's spectacular restrooms and Apple's emotional design pull attention to attributes rivals ignore, causing customers to re-rank what matters.

Choose which attribute to make surprising — attention rewrites the buyer's priorities.

Principle

Two ways to make money: make desirable things or make things desirable

Making things desirable is as profitable as making desirable things, and far more malleable.

Most operators only pull the product lever. Sutherland argues the psychology lever is equally profitable and less constrained because the laws of psychology, unlike the laws of physics, can be bent.

Before spending to improve the product, ask whether the same money spent on perception would move behavior more.

Principle

Manufactured signals of progress trigger dopamine and improve the experience

Showing progress during a wait changes the felt experience through intermittent reward.

Domino's tracker and pizza 'in the oven' status build felt progress; McDonald's kiosk that only shows 'in preparation / ready' misses the dopamine of forward movement.

Give users visible forward movement during any wait.

Principle

Capture the upside with outcome-based pricing

Price on the value you create, not the cost of the box, to capture and align on the upside.

Watt and Boulton took a third of coal savings; the model self-selected into Cornwall where coal was priciest, maximizing both customer savings and their cut. Rolls-Royce later priced jet engines by the hour.

Structure pricing so you profit precisely when and where the customer profits most.

Principle

Marketing and innovation are fat-tailed activities

A minority of bets carries the value, and you can't identify them in advance, so you must keep exploring.

Bezos: in baseball the most you score is four, but in business you can hit a thousand. Killing exploration to hit reliable singles forfeits the grand slams.

Protect a discovery layer of unjustifiable bets; efficiency drives kill the fat tail.

Principle

Value is produced in the mind, not the factory

Value lives in perception; optimizing only production ignores where value is actually created.

As firms scale they fixate on what they do rather than what they mean, losing the meaning-making that generates perceived value.

Manage the meaning of the product, not just its manufacture.

Principle

Reduce anxiety rather than increase range

Attack the emotional symptom directly instead of the costly physical proxy for it.

Chasing range to kill anxiety yields heavier, pricier cars with huge idle batteries; reframing the gauge would do it for free.

Find the anxiety behind the metric and solve the anxiety.

Principle

An invention isn't an innovation until it changes behavior

If it doesn't change behavior, it's an invention, not an innovation.

The industrial revolution was a marketing revolution as much as an industrial one — abundance is worthless without corresponding demand.

Measure your innovation by behavior changed, not features shipped.

Principle

The laws of psychology are more malleable than the laws of physics

Businesses overspend fighting immutable physics when malleable psychology is the cheaper win.

The EV range-anxiety example: reframing a 16% reading as 58 remaining miles removes panic without adding a kilowatt-hour of battery.

When a hard engineering problem is really a perception problem, solve the perception.

Principle

All data comes from the past, so data-only decisions inherit a status-quo bias

Data-only decision-making converges you onto your competitors and away from the future.

Everyone focusing on the same measurable metrics becomes indistinguishable — 'red water competition'. Red Bull could only be justified on intuition against all the rational data.

Use data to sense-check intuition, not to replace it — the future has no dataset.

Principle

Compete on the neglected dimension, not the obvious one

The biggest win is usually in the unmeasured step next to the core product.

Given 'make a better taxi', everyone optimizes the ride; Uber fixed booking/waiting/paying and tripled the San Francisco taxi market rather than merely taking share.

Ask what surrounds the core experience — the bottleneck is usually there.

Frameworks

Reusable systems and operating models — including when they help and when they break.

Framework

Test one thing or test everything — never five things at once

Either isolate a single variable or swap the entire paradigm — mixing several changes destroys attribution.

Direct-response copywriters tested single words ('do you make mistakes' vs 'these mistakes' vs 'these common mistakes') because tiny changes drive butterfly-effect differences in response.

Design experiments as one-variable or whole-new-concept, never a muddle in between.

Framework

Reverse benchmarking: find the neglected metric and double down on it

Enumerate the metrics everyone competes on, pick the one all have neglected, and be spectacular at it.

From Unreasonable Hospitality: don't copy the world's best restaurant, find what it does badly (coffee, beer service) and hit that out of the park. Buc-ee's did it with restrooms, Apple with feel.

Audit the category's ignored metrics and over-invest in one.

Framework

Ask what business you are really in — the meta-need behind the purchase

Reframe the product around the unspoken emotional job it actually performs.

Requires letting go of two handholds at once: economic logic (lower price sells more) and conventional market research (customers will tell you what they want).

Interrogate the meta-need: what is the customer really buying?

Signals

What appears to be shifting, for whom it matters, and what happens if you ignore it.

Signal

Discarded ad forms that still work are an arbitrage — e.g. comic-strip ads meet manga

Techniques abandoned as unfashionable often never stopped working — the archive is arbitrage.

Long copy, direct mail, jingles and cartoon-strip storytelling never stopped working; comic-strip ads map onto manga to reach the young. You advertise to a moving parade, not a standing army.

Do advertising archaeology — revive proven forms the industry dropped for fashion.

Signal

AI products need a 'Horsepower' — a legible marketing unit instead of confusing benchmarks

The AI category is wide open for a legible outcome-unit that ends benchmark confusion.

Clawed, ChatGPT and rivals ship near-identical chat interfaces and name tiers (light/heavy/pro/plus) plus arcane coding benchmarks buyers can't parse — a Horsepower-shaped gap.

Invent the customer-legible unit for your category before a rival does.

Opportunities

Only included where there is a buyer, a real wedge, and a plausible revenue path — not vague idea theater.

Opportunity

Age-beneficial tech is under-marketed to the old who would value it most

There is unclaimed demand in reaching older buyers that youth-obsessed marketing ignores.

Bone-conduction headphones (bypass the deteriorating inner-ear bones) and Samsung folding phones (bigger screen for failing eyesight) are godsends to the old but marketed only to the young.

Build the go-to-market that reaches valuable older buyers rivals won't address.

Opportunity

Flat White or Off: pre-made premium coffee for time-pressured venues

Strip customization to maximize throughput of a high-quality default where speed matters most.

'Henry Ford applied to coffee' — two versions (oat/regular), tap and go. The best-discovered market was conferences, where customizable queues bottleneck at every break.

Where speed dominates choice, sell one great default fast.

Lessons still worth keeping

Useful takeaways that did not fully clear the bar for durable principle status.

Lesson

Fit beats self-judgment: the same person fails in one role and tops another

Talent is role-relative; diagnose fit rather than labelling people good or bad.

Sutherland was fired from a non-copywriting role, returned to direct-response copywriting and rose to the top — a lesson in matching nature to role.

When someone fails, re-examine role fit before writing them off.

Lesson

Direct response was behavioral economics decades before it had a name

The old direct-response canon already contains empirically-validated behavioral tactics.

Copywriters ran randomized trials via interleaved newspaper presses and measured coupon codes in the 1920s-50s, learning that tiny changes produce butterfly-effect differences in behavior.

Mine pre-behavioral-economics direct-response literature for tested, still-working tactics.

The Plays

Try these this week

Verb-first executable actions — each one tied to a stated outcome in the episode.

Fix the pre-product experience rivals never measure (airport rental reframe)

Outcome: Invest in the unmeasured pre-product moment customers actually hate.

Context: Sutherland books taxis to hotels then rents downtown next day to skip the airport-with-luggage ordeal. A meet-at-arrivals, luggage-carried, car-pre-cooled service would command a premium.

in some cases I would literally happily pay a hundred quid. If the deal is we meet you at arrivals, okay? We take the luggage to the car, we hand you the keys, we answer your questions about how the car works. If by the way, they also turn the air... conditioning on 10 minutes in advance.
Rory Sutherland
one to two quarters per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · control over the entry experience

Offer redundant response channels to unblock the psychological bottleneck

Outcome: Offering multiple response channels unblocks buyers stopped by channel-specific friction.

Context: A randomized 3-way phone-company test (50k each): post-only 5%, phone-only 2%, choice-of-both ~7% — proving how you can order matters more to conversion than what or how much.

post only 5% response rate, phone only 2% response rate. When you offer people a choice, it was just, it was 7%, not quite, but almost nearly, I think the sum total of the, of the two independent response rates
Rory Sutherland
per test cycle per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · ability to run randomized splits and attribute response

Name the trade-off in the product itself so it is accepted before arrival

Outcome: Bake the trade-off into the name so customers self-select and never feel short-changed.

Context: Flat White or Off, the Slate truck ('crank the window yourself'), Moxy hotels (tiny rooms, no room service), and Avis 'We're number two, we try harder' all make the trade explicit up front.

the reason it's called flat white or off is it makes explicit the trade off before someone turns up. So you said if anybody, the joke is, if anybody turns up and asks for a cappuccino, you point at the logo and you say the answers in the name.
Rory Sutherland
set at launch per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · a segment that values the thing you keep over the thing you drop

Invent a marketing unit that maps your product to the customer's real calculus

Outcome: Coin a unit denominated in what the customer actually cares about.

Context: Boulton asked what buyers needed to know before buying — how many horses they could stop feeding — and invented Horsepower. Apple did it with '1,000 songs in your pocket'.

what went out and invented a unit we still use today, which is called the Horsepower... the reason it's not named after a famous scientist... is because it's a marketing unit. It was invented for marketing purposes because he could then go and say if you buy 25 Horsepower steam engine, you can actually get rid of 75 horses.
Rory Sutherland
weeks to define, permanent once adopted per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · clear view of the buyer's cost or benefit calculus

Sell hardware as a service: give it free, take a share of the savings

Outcome: Zero upfront, share-of-savings pricing removes buyer risk and captures the upside.

Context: Watt & Boulton took a third of coal savings; installs concentrated in high-coal-cost Cornwall, maximizing both parties. Rolls-Royce later charged airlines per engine-hour.

They said, we'll supply it for free, you pay us a third of the money you save on coal... it aligned the interests of the people selling the steam engines and the people owning the mines because the people in the country where coal was most expensive, most needed to make mo to save money on coal.
Rory Sutherland
contract term (multi-year) per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · measurable outcome
  • · trust/audit mechanism
  • · balance sheet to fund upfront

Appoint a champion for the category's worst-served detail

Outcome: Give a dedicated owner the mandate to be spectacular at the neglected detail.

Context: From Unreasonable Hospitality: after learning the world's best restaurant treated beer drinkers shabbily and served average coffee, Will Guidara appointed a coffee and a beer sommelier to excel there.

he goes back to his own restaurant and he appoints a coffee sommelier and a beer sommelier and he says, your job is not just to benchmark against these people, it's to hit it out of the park.
Rory Sutherland
one quarter to stand up per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · willingness to over-invest where rivals coast

Reframe 'apply' as 'receive' to remove fear-of-rejection friction

Outcome: Convert 'apply' into 'receive a pre-approved offer' to remove ego risk from conversion.

Context: People already wanted the AmEx Gold card; the blocker was fear of rejection, so the letter said 'sign to receive', not 'apply'. Difference: Yale inviting you to apply vs Yale saying they want you.

he managed to find a way in which you could actually say, all you need do to receive your card, not apply for your card, but receive your card is sign your name on this form, what we know about you suggest you're virtually pre-approved for membership.
Rory Sutherland
per campaign per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · ability to pre-qualify
  • · product people already want

Manufacture visible forward-movement during any wait

Outcome: Show staged progress during waits to supply the dopamine of forward motion.

Context: Domino's tracker and pizza 'in the oven' status work this way; McDonald's kiosks that show only 'in preparation / ready' forfeit the effect. Sutherland would 'defend in court' the practice as making people feel better.

they could probably play a game there where they say, you know, being prepared... on the tray they could give you an impression of forward movement because it's all to do with dopamine and, and sort of effort and reward.
Rory Sutherland
days to implement per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · some signal of actual progress to anchor the display

Add a decoy option to steer choice toward the target tier

Outcome: A dominated decoy reframes the target tier as a free upgrade and lifts its take-rate.

Context: The Economist's paper-only option (same price as digital-only) made paper+digital look free-with-paper; removing it collapsed premium subscriptions. Valuable because print circulation commands higher ad revenue.

the presence of the middle option increased by about two or 300%. The number of people who subscribed to Paper plus digital instead of digital
Rory Sutherland
per pricing test per
  1. 1

  2. 2

  3. 3

  4. 4

Scripts

Before you start

  • · a target tier and a plausible dominated decoy

Decision Moments

Actual decisions, real outcomes

Specific decisions narrated in the episode with their outcomes and transferable lessons.

Watt & Boulton were selling more-efficient steam engines to mine owners who already ran cheaper, less efficient Newcomen engines, and a flat sticker price captured none of the value the new engine created.

Did: Instead of charging a fixed price (~a hundred guineas), they supplied the engine free and charged one third of the coal the owner saved — hardware as a service in 1775 — which self-selected installs into high-coal-cost Cornwall where savings, and thus their cut, were largest.Outcome: They aligned incentives with mine owners and captured a proportional share of the upside they created, concentrating deployment where value was highest; Rolls-Royce later mirrored it with per-hour jet-engine pricing.

Price on the value delivered, not the cost of the box, to capture upside and align incentives.

Part of an emerging decision pattern across multiple episodes

A phone-company client (c.1991) wanted to drop the postal coupon response entirely and offer only phone ordering for paid 'star services', on a hunch, risking the business if the channel mattered.

Did: Rather than cut the channel on a whim, they ran a randomized test of three letters to 50,000 people each: post-only, phone-only, and choice-of-both.Outcome: Post-only pulled 5%, phone-only 2%, and offering the choice pulled ~7% — nearly the sum of the two — proving the ordering channel mattered more to conversion than the product or its price, and that dropping post would have roughly halved response.

Test channel and framing changes rigorously before acting on a hunch; psychological bottlenecks in how people order can dwarf product and price.

Part of an emerging decision pattern across multiple episodes

In the late 1980s/early 1990s American Express pushed the pricey Gold card with long letters listing rational reasons to apply, but uptake was limited by applicants' fear of rejection rather than lack of desire.

Did: On a client's intuition that people already wanted a gold card, they reframed the offer so recipients need only sign to 'receive' — not apply for — a card they were 'virtually pre-approved' for, removing the ego risk of rejection.Outcome: Reframing 'apply' as 'receive a pre-approved offer' unlocked buyers who already wanted the product but feared rejection; paired with the near-free 'member since' date, tiny psychological details drove outsized retention and acquisition value.

When desire exists but conversion lags, diagnose fear of rejection and reframe the ask as acceptance of a pre-approved offer.

Part of an emerging decision pattern across multiple episodes

Tensions surfaced

Contradictions and trade-offs the episode raises — judgment calls a thoughtful operator has to navigate.

Tension

Genuine welfare vs manipulation in placebo-style psychological design

Placebo-style design can raise real welfare while trading on illusions users would object to.

Placebo elevator door-close buttons give 'the illusion of agency'; branded painkillers that are chemically identical still relieve more pain. Sutherland is candid it is 'mind hacking' yet defends it as making people feel better.

Weigh whether a psychological lever genuinely helps the user or merely exploits them.

Tension

Defensibility of a decision vs quality of its outcome

Optimizing for defensible decisions degrades outcome quality by forbidding intuition.

Like a detective allowed only evidential value and no hunches, an org that requires every decision be defensible reaches only the crowded, already-occupied fifth of the solution space.

Notice when your process is optimizing defensibility over outcomes, and protect space for intuition.

Corpus connection

Where this episode fits for retrieval

What kinds of decisions this briefing is best pulled into.

Primary decisions

  • positioning
  • pricing
  • go-to-market