营销用错了数学：Rory Sutherland 谈好产品为何还需要心理、时机和名声

Rory Sutherland opens by warning product builders not to assume that good products automatically succeed or that bad ones necessarily fail. He adds that timing matters so much that people should not reject an idea simply because it failed in the past. He cites Meta Portal TV as one of the best products he has worked on: he bought about four of them, considered it roughly $500 worth of equipment sold for about $120, and bought another when it was discontinued. Reviews called it brilliant as a product, but the first seven or nine paragraphs focused on privacy paranoia about allowing Facebook to put a camera in the home.

Sutherland uses the Portal TV case to show how a product can be judged before its actual utility is discussed. He notes that video calling on television still is not common unless people plug in a laptop or do something complex. He then quotes his own warning that the urge to appear serious is in many ways a disaster in marketing, and that if a stand-up comedian could do a routine about a product, the team is onto something.

Sutherland invokes the French designer Raymond Loewy and the concept of MAYA: maximally advanced yet acceptable. He argues that consumers are generally more comfortable with evolution than with complete reinvention, and that there is a pace of change they will accept. Even the iPhone was preceded by the iPod, so it was not a complete 'what the hell' moment; consumers like to migrate their behavior rather than reinvent it.

His rule is that idiosyncrasies count double. When Ogilvy won the Jaguar account in the 1990s, a creative director said the company needed to preserve slightly odd things, such as a reading-light switch located on the central console rather than overhead. He also cites Rolls-Royce's floor pedal for dipping headlamps and the DoubleTree cookie at check-in as small details that garner disproportionate attention.

Sutherland points to MCI's friends-and-family plan, where customers nominated frequently called numbers and received a discount on those calls. The plan was somewhat irrational but generated more interest than a flat 15-20% reduction because customers had to stipulate their numbers. He also tells how Veuve Clicquot's yellow label began as a printing mistake and became the champagne's visual identity, and how BBC 4 bought The Killing for £24,000, about $30,000, before it became a hit because Scandinavians were 'just the right amount of weird'.

Sutherland describes psychology as a branch of complexity theory, full of nonlinear butterfly effects where small things have huge effects. He argues that the opposite of a good idea can be another good idea. He gives two opposite but equally great hotel check-ins: an insanely high-touch experience at the Mandarin Oriental in Hong Kong, and a no-touch experience where a guest walks in with a phone and unlocks the room.

He argues that business tries to run on a reductionist maths and physics model where everything is linear, proportionate and has a single right answer. He criticizes intelligence tests and the SAT for being multiple choice, since real decisions can have multiple acceptable answers and may not contain all necessary information. In real life, some information in the problem is irrelevant and some needed information is missing.

Sutherland recalls a French experiment in which people were told a ship stops, 27 goats get on, three sheep get off and 25 cows are added, then asked how old the captain is. Even intelligent children assumed the information must be useful and did not answer that it was impossible to tell. He says business decisions involving human behavior do not meet the conditions of scientific questions, yet people are selected and promoted on their ability to perform those artificial activities.

Sutherland highlights companies that give small teams autonomy within a clear overall objective. He cites Octopus Energy and its software division Kraken, where small autonomous teams of roughly 10 to 20 people operate within a very clear brief and offer tariffs that make charging an electric car after midnight much cheaper. He also cites Shopify, where the COO modeled customer service teams of 10 on sports teams; the structure was less lean, but people were highly motivated and covered for teammates.

He connects this to the Dunbar number of about 150 people, the number of people one can join in conversation without feeling awkward. In teams of 10, if one person takes a sick day, the other nine feel the obligation directly, whereas in an organization of 100 that reciprocity does not scale. He also describes AO Appliances Online, whose briefs include 'treat the customer like you treat your grandmother' and 'would your mum be proud of what you've done?', and Zappos, which refused to make call speed a metric and had one customer service call last seven hours.

Sutherland warns that metrics often get gamed or obeyed while being found stupid. In a call center incentivized to get customers off the phone quickly, employees lose autonomy and judgment, which he calls deeply depressing and demotivating. He describes quarterly target obsession as 'Soviet-style capitalism' and argues that persistent cost-cutting has destroyed much of the pleasure of work by removing discretion. He also references John Ralston Saul's Voltaire's Bastards, which warns against fixating on reason to the exclusion of ethics, memory, instinct and creativity.

Sutherland argues that the ability to behave irrationally can be rational at the meta level. He says there are pages in his book suggesting that being a bit Donald Trump has its virtues, because no one is entirely sure what you will do, while adding that he is far from an uncritical admirer. In New York real estate, being completely predictable is not necessarily an advantage. If you never fight back against anyone slightly taller, you may spend your life being pushed around by people a quarter of an inch taller.

He draws on evolutionary biology to explain why randomness has a function. When a hare is chased by a dog, it enters a random, high-speed pattern of movement, much like a Russian submarine executing a 'crazy Ivan'. The hare cannot consciously know its next move, because knowing it would reveal preparatory muscular movements the dog could read.

Sutherland says humans do not utility-maximize; they rely on habit and social copying. An organism that learned everything from first principles would eat too many poisonous berries, so copying past behavior or the behavior of others is a low-cost way to discover low-downside behavior. He notes research on heat pumps finding that knowing three people on your street had one was more persuasive than environmental benefits or savings, and he says electric-car buyers rationally wait because electrical products historically get better and cheaper quickly.

Sutherland jokes that there are no famous Belgians because people assume they are French, just as famous Canadians are assumed to be American. He says the same happens with marketing: when a product succeeds, people forget that marketing was instrumental and assume the product was obviously great. The Ford Model T and the iPhone look inevitable in hindsight, but electricity in the home was advertised for about 30 years, dial-up internet had to be sold in the late 1990s, and the mobile phone was partly driven by social pressure.

He cites Edward Jenner's smallpox vaccination, which faced massive skepticism and suspicion. Jenner's marketing coup was getting the British royal family to vaccinate their children. Sutherland also argues that Steve Jobs was not a technologist but a brilliant pitch man and marketer; Apple's technical people did not respect him, but he excelled at product design, focus, taste and presentations that made people believe Apple could do magic.

Sutherland warns that we only look at successes, creating survivorship bias. He lists Meta Portal TV, Google Glass, the wine box, the Japanese toilet and, until recently, the air fryer as brilliant products that failed to reach critical mass because of timing or marketing rather than the product itself. The wine box keeps for weeks and has its own tap, and the Japanese toilet is superior to dry paper, yet neither became the Western default. He also points out that people still open houses with metal keys while cars open easily, and that the Romans never invented the stirrup.

He insists that past failure does not mean an idea should not be tried again. The definition of insanity is not trying the same thing repeatedly; that is often the definition of a complex system, and many internet ideas failed because they were mistimed. He mentions the Chumby, a small internet-connected device he owned about three of, and recalls being abused from passing cars in 1989 for using a mobile phone on Oxford Street. He also remembers being laughed at in the early 2000s for ordering groceries online, and later fearing mild physical attack for making an international call on a mobile phone from the top of a London bus.

Sutherland cites Herodotus on the ancient Persians, who debated matters twice, once sober and once drunk, and proceeded only if they agreed in both states. He sees this as a two-stage, double-lock decision process that combines rational and emotional appeal. He notes a fundamental asymmetry: creative people must present ideas to rational people for approval, but engineers and accountants never present their numbers to wacky people to see if there is a neater idea. He also describes Ford and Edison's early electric car, which was quiet, had good acceleration and decent range for the time, but was killed by user imagery because it became stereotyped as a women's car while farm boys used gasoline cars.

Fast-forwarding to today, Sutherland says electric cars are lovely to drive, quiet and have few moving parts, yet many people hate them for reasons that boil down to user imagery. The first adopters of innovative products tend to be slightly weird, and weird people do not confer the reassurance of social norms. Non-owners see electric-car owners as smug environmentalists looking down on diesel drivers, even though many owners are not especially interested in the environment. He notes that charging points were plentiful in less fashionably progressive places and scarce in places like Cambridge and Brighton.

Sutherland argues that much technology adoption is driven by status display, like a peacock's tail. The first cars were not as good as horses and carts but were used for novelty and showing off, and rich people provide early-stage funding for promising ideas. He compares this to birds evolving wings as sexual display plumage before they became large enough for propulsion. He also describes his air-fryer evangelism, which made people see air-fryer owners as members of a cult rather than as reassurance, and warns marketers that they are high in openness to experience while most people are driven by habit and social norms.

Sutherland tells the story of the Walkman and Sony's Akio Morita, who rejected adding recording functionality even though it would have cost only about 50 cents or a dollar. Morita wanted the device to have one function performed very well, with no ambiguity about what it was for. He contrasts this with McDonald's choice reduction versus the customization of the American diner, and with Tesla's choice architecture, which offers just the right amount of choice. Customizing a new electric Range Rover, by contrast, drives people insane because the options quickly add up to the price of a house.

For startups, Sutherland's brand advice is simple: be consistent, be distinctive and be famous, with some clarity of promise. He says advertising agencies avoid saying 'we're going to make you famous' because it sounds too obvious, but fame fundamentally changes the rules in non-linear ways. When you are not famous, you have to find all your customers; when you reach escape velocity, people start coming to you and suggesting applications you had not considered. Fame cannot easily be attributed to one event, and consumers often do not know how they heard about a product.

Fame means calls get returned, people come to work for you for less money, people apply without being found, customers give you the benefit of the doubt, and you are allowed to cock up once. Sutherland quotes the definition that a great brand lets you play the game of capitalism on easy mode, because you cannot know what your score would have been on hard mode. He points to Coke's level of fame, where customers expect any shop or bar to stock it, and to B2B reassurance: nobody got fired for buying IBM. He also notes that people would rather explain a BA delay than a Ryanair delay, because brands are extended phenotypes that signal what using them means about the user.

His final advice is to do the psychological, technological and economic work in parallel rather than in series. The sweet spot is something people want psychologically, something that works technologically, and something that makes money economically. Businesses borrow too much from a Fordist, Taylorist production-line mentality and pretend innovation is linear, but the real process is not. He cites Wrigley starting with soap powder, giving away baking powder, then chewing gum, and becoming a chewing gum company, and he argues that marketing and innovation are two sides of the same coin: either work out what people want and find a clever way to make it, or work out what you can make and find a clever way to make people want it.