Rory Sutherland, an advertising executive at the agency Ogilvy, argues that the modern world's fixation on logic and economic rationality has blinded decision-makers to a powerful category of solutions he calls "alchemy": the use of psychology, perception, and seemingly irrational thinking to solve problems. Drawing on behavioral economics, evolutionary psychology, and advertising case studies, Sutherland contends that human behavior follows its own evolved rules, which he terms "psycho-logic," and that these rules diverge sharply from the narrow rationality assumed by economists and technocrats.
Sutherland opens with the paradox of Red Bull. A logical approach to challenging Coca-Cola would produce a drink that tastes better, costs less, and comes in a large bottle. Red Bull did the opposite: It was expensive, came in a tiny can, and received the worst consumer taste-test results the research agency had ever seen. Yet it sells 6 billion cans annually. This establishes the book's central premise: Alongside the proven achievements of science and logic, hundreds of seemingly irrational solutions to human problems await discovery by anyone willing to abandon conventional thinking.
In the Introduction, Sutherland presents a charity fundraising experiment. The "rational" envelope promoting a tax rebate reduced donations by over 30 percent, while three seemingly irrelevant changes (higher-quality paper, portrait orientation, mention of volunteer delivery) each increased donations by more than 10 percent. He introduces four reasons humans behave in seemingly illogical ways: signalling, or costly actions that convey trustworthiness; subconscious hacking, or indirect methods of influencing one's own unconscious states; satisficing, or choosing good-enough options under uncertainty; and psychophysics, or the ways perceived experience diverges from objective reality. These categories structure the rest of the book.
Part 1, "On the Uses and Abuses of Reason," argues that reason is overused as a problem-solving tool. Sutherland introduces the "broken binoculars" metaphor: Market research and economic theory are two cracked lenses through which decisions are viewed, both providing distorted pictures of human motivation. He illustrates psycho-logical thinking through an energy company whose customers complained about waiting all day for heating engineers. Rather than creating expensive one-hour appointment windows, a text message sent 30 minutes before arrival could eliminate most emotional distress at a fraction of the cost, because the real source of frustration was uncertainty, not the wait itself. He extends this to "psychological moonshots," arguing that making a train journey 20 percent more enjoyable may cost almost nothing, while making it 20 percent faster costs hundreds of millions. He also introduces physicist Ole Peters's distinction between ensemble and time-series perspectives, arguing that many supposed cognitive biases may reflect rational behavior when viewed from the perspective of how real life is actually lived.
Part 2, "An Alchemist's Tale," contends that psychological value creation is real and that organizations should actively pursue it. Sutherland tells the story of Frederick the Great, who failed to convince Prussian peasants to cultivate potatoes through coercion. Instead, Frederick declared the potato a royal vegetable, posted guards around a royal potato patch with secret instructions not to guard too closely, and let peasants "steal" the exclusive crop. He discusses "the alchemy of semantics," showing how the Patagonian toothfish became the Chilean sea bass, a $20 restaurant dish, and how Harvey Mudd College in California quadrupled its female computer science majors by renaming courses, dividing classes by experience level, and addressing classroom culture rather than imposing quotas. He also argues that reducing features can increase appeal, citing Sony co-founder Akio Morita's decision to remove the recording function from the Walkman because multiple functions confused consumers about the device's purpose.
Part 3, "Signalling," explains how costly investments of money, effort, or time convey trustworthiness in ways cheap talk cannot. Sutherland uses the London black cab Knowledge, a grueling four-year program requiring memorization of 25,000 streets and 20,000 landmarks, to illustrate how costly initiation deters cheats and builds trust. He introduces "continuation probability," the prospect of repeated interaction, as the primary mechanism underpinning trust. Through an analogy with bees and flowers, he explains branding: Flowers invest in distinctive petals and scent to attract bees, creating a feedback loop that rewards quality. Without distinctive identity, a flower offering extra nectar gains nothing because bees cannot learn to prefer it. Brands serve the same function for humans. Sutherland extends the framework through Charles Darwin's theory of sexual selection and biologist Amotz Zahavi's handicap principle, which proposes that costly traits like the peacock's tail signal genetic quality precisely because they are expensive to maintain.
Part 4, "Subconscious Hacking," argues that humans use oblique methods to influence unconscious processes beyond direct conscious control. Drawing on psychologist Nicholas Humphrey's theory that the immune system has a switch influenced by the mind, Sutherland proposes that placebos work by convincing the unconscious that conditions are favorable for healing. He extends this beyond medicine: Military paraphernalia functions as a "bravery placebo," and consumer spending on cosmetics and luxury goods serves as what social psychologist Jonathan Haidt calls "self-placebbing," in which people self-administer confidence they cannot generate consciously. A 2017 study from INSEAD and the University of Michigan demonstrated this principle with Red Bull, finding that men who were told their identical cocktail contained Red Bull reported feeling drunker and behaved more confidently, showing the effect was perceptual rather than chemical.
Part 5, "Satisficing," addresses why humans make good-enough decisions under uncertainty. Sutherland draws on economist Herbert Simon's concept, which combines "satisfy" and "suffice," arguing that seeking satisfactory solutions under incomplete information is often smarter than pursuing perfect ones. He uses the analogy of rogue bees that ignore the hive's waggle dance to explore at random: Without these inefficient scouts, the hive would exploit existing food sources until they ran out. Brand preference is itself a satisficing strategy, since people choose a brand not because they believe it is the best but because they are more certain it is acceptable.
Part 6, "Psychophysics," examines how perception differs from objective reality. Sutherland notes that Cadbury's received complaints about changing the taste of their chocolate when the company had only changed the shape of the blocks, because smoother shapes taste sweeter. He shows how language alters experience: A Belgian biscuit manufacturer lost sales after printing "Now with lower fat" on packaging of a product that tasted identical in blind tests. He also describes the IKEA effect, whereby General Mills' Betty Crocker cake mixes failed until the instructions required adding a real egg, because a small effort increased perceived value.
In Part 7 and the conclusion, Sutherland distills practical lessons, including the story of a retail website that increased annual sales by $300 million by replacing a "Register" button with a "Continue as Guest" option. He discusses economists Richard Thaler and Shlomo Benartzi's "Save More Tomorrow" pension scheme, which links contributions to future pay rises so contributors never feel poorer. He notes a structural reason organizations resist psychological solutions: Logical failure is forgivable, while imaginative failure gets the decision-maker fired. Sutherland warns against what he calls "the arithmocracy," the class of data-focused decision-makers whose quantitative lens prevents consideration of psychological alternatives. He closes with the metaphor of "scenting the soap": Effective solutions must appeal to unconscious motivations, not just rational ones, because the apparently pointless feature is often what makes the whole system work.