Rory Sutherland's marketing ideas, and how to use them

Rory Sutherland is the vice chairman of Ogilvy in the United Kingdom and the best known popular advocate for applying behavioural science to marketing. His central argument is simple and awkward for most businesses: value is created in the mind of the customer, so a psychological change can be worth as much as an engineering one and usually costs far less. A train journey can be improved by making it faster, which is expensive, or by making the time spent feel better used, which is cheap. Both are real improvements. Because only one of them appears on a spreadsheet, companies systematically overspend on the measurable and underspend on the perceptual. His other recurring theme follows from that: tests of small, strange, unmeasurable things often produce the largest returns, and the reason they go untried is organisational rather than commercial.

This page sets out the ideas and turns them into things a marketing team can actually do. It is a practical summary of a body of public work, including his book Alchemy, his talks and his columns, not a biography.

The six ideas worth taking

1. Perceived value is real value. If a customer is more satisfied, waits more happily, or trusts you more, you have created value whether or not the product changed. The practical consequence is that the experience around a product deserves a line in the budget, not just the product itself.

2. Psychological solutions compete with logistical ones. Before funding a costly operational fix, ask what a change in framing, sequencing or information would achieve. Telling a customer accurately when something will arrive often does more for satisfaction than making it arrive sooner.

3. The opposite of a good idea can also be a good idea. Business logic treats strategy as a single optimum. In reality both a cheap self service option and an expensive attentive one can work, while the compromise in the middle appeals to nobody. This is an argument against averaging your positioning.

4. Costly signalling explains a lot of marketing. Expensive, visible, hard to fake effort is believable precisely because it is expensive. A substantial guarantee, a long warranty, a physical showroom or a well made brochure all say something a claim cannot, because a weak competitor could not afford to copy it.

5. People choose from a shortlist, not from the market. Buyers satisfice. They pick something acceptable from the few options they already trust, under time pressure, with imperfect information. Getting onto the shortlist is a different and often cheaper job than being judged best.

6. Measurability bias distorts budgets. What is easy to count gets funded and what is hard to count gets cut, regardless of effect. The remedy is not to abandon measurement but to be explicit when an activity is being judged on an incomplete measure.

The difference between this and ordinary persuasion advice

Behavioural science in marketing is sometimes reduced to a list of biases to deploy on a landing page. The more useful reading is about where to look for opportunities rather than which trick to apply. Three implications are specific:

• Context beats copy. The same offer performs differently by order, default, comparison set and timing. Changing the context around a choice is usually cheaper than changing the offer.
• Reasons matter as much as facts. A plausible reason attached to a claim makes it credible. "Our coffee costs more because the beans are roasted here each morning" works where "premium coffee" does not.
• Friction is a tool with two edges. Removing friction raises conversion. Adding it in the right place, such as a short qualifying question, raises quality and can raise perceived value.

Turning the ideas into work

Instead of	Try
Spending to make a process faster	Spending to make the wait informed and bearable, with accurate progress information
A discount to win a hesitant buyer	Removing the risk: a guarantee, a trial, a reversible first step
Describing your product in category language	Describing the specific moment it is for, which is what makes a reader recognise themselves
One middle positioned offer	Two offers at opposite ends, each clearly for a different buyer
Claiming quality	Showing a costly signal of it: a long guarantee, published method, named people, a physical presence
Optimising the last step of the funnel	Working on whether you are on the shortlist at all, which is where most of the loss happens

How to test ideas like this without losing the argument

The common objection is that perceptual work is unmeasurable. It is measurable, with a little care:

• Pick one surface where you have volume, such as a confirmation email, a booking step, a pricing page or a quote document.
• Change the frame, not the facts. Same price, same product, different reason, order, default or comparison.
• Measure a behaviour, not a sentiment. Completion, upgrade, reply, retention at ninety days.
• Run it long enough to clear the novelty, and keep the control running.
• Write down the prediction first. Being wrong in writing is how a team earns the right to try stranger things.
• Keep a log of what you tried, including failures. The value compounds across a year.

A sensible cadence is one framing test per quarter on a high volume surface, alongside your ordinary conversion work. The point of the log is that it gives you an evidence base for the next unmeasurable looking idea.

Where to read the primary material

His book Alchemy: The Dark Art and Curious Science of Creating Magic in Brands, Business and Life is the fullest statement of the argument, and the earlier collection The Wiki Man gathers his columns. He speaks frequently at industry conferences and in podcasts, and much of that is public. For the research underneath the ideas, the work of Daniel Kahneman on judgement, Richard Thaler and Cass Sunstein on choice architecture, Robert Cialdini on influence, and the marketing effectiveness research of Les Binet and Peter Field on the balance of brand and activation all sit alongside it.

The Rory Sutherland's marketing ideas questions worth an answer

What is Rory Sutherland best known for arguing?

That value is psychological as well as material, so a cheap change in perception can be worth more than an expensive change in engineering, and that businesses systematically miss these opportunities because the gains are harder to put in a spreadsheet than the costs.

Is behavioural economics in marketing just manipulation?

The useful version is not. Making a wait informative, offering a reversible first step, giving an honest reason for a price and removing a pointless form field all improve the customer's experience as well as your numbers. Techniques that depend on the customer being misled fail on repeat purchase and damage the brand, which is the commercial argument against them as well as the ethical one.

How does this apply to a small business?

Better than it applies to a large one, because you can change the frame on Monday. Rewrite the quote document to give reasons for the price, add a guarantee that a competitor would not dare copy, say precisely when you will call back and then do it, and describe the exact situation your service is for rather than the category you are in.

What is the practical difference between brand and performance marketing in this view?

Brand work builds the shortlist you are chosen from. Performance work harvests demand from people already looking. Both are needed, and the measurability of the second is the reason it tends to be overfunded relative to the first.

Where should a team start?

With the single highest volume moment where a customer forms an impression of you, which is often the quote, the booking confirmation or the first reply to an enquiry. Improve the framing there, measure a behaviour, and use the result to fund the next experiment.

A new place the shortlist gets formed

The shortlist idea has a literal and recent version. When somebody asks ChatGPT, Gemini, Perplexity, Claude or Google's AI Overviews which company to use for something, the answer names two or three. That is a shortlist, assembled without the buyer comparing anyone, and each assistant builds it differently from different sources.

It is worth knowing whether you are on it in your own category. Ask the question a customer would ask, keep your own brand name out of the wording so the answer is not steered, and note who gets named and which pages get cited. AI Knows Us runs that check across those assistants and shows the sources behind each answer.