Failed Forward

The idea began with a household annoyance: in the late 1970s, Dyson's expensive new vacuum kept losing suction as its bag clogged. Visiting a local sawmill, he watched an industrial cyclone separate sawdust from air using nothing but spinning force — no bag, no clogging — and asked the question that would consume fifteen years: why doesn't a vacuum cleaner work like that?

The middle of this story is what nobody envies. Working out of the coach house behind his home, Dyson built prototype after prototype — cardboard, brass, one variable changed at a time, the slow way, the honest way. The family lived lean on his wife Deirdre's earnings as an art teacher while the count climbed into the thousands. He has been precise about the number ever since, because the number is the point: five thousand one hundred and twenty-six prototypes failed. The five thousand one hundred and twenty-seventh worked.

Then came a second wall he hadn't planned for: the industry said no. The major manufacturers he approached had little interest in a bagless vacuum — replacement bags were a rich, recurring revenue stream, and his invention would kill it. The companies whose products he'd improved upon declined to make the improvement. So he licensed the design in Japan, where it sold as a sleek, expensive cult product — and used those royalties to do the thing inventors are told not to do: he opened his own factory.

The DC01 launched in 1993 and became the best-selling vacuum in the United Kingdom within about eighteen months. Dyson the company went on to reinvent hand dryers, fans, and hair dryers — and Dyson the man became one of Britain's wealthiest people, still telling anyone who asks that the failures weren't the obstacle to the invention. They were the invention: each dead prototype eliminated a wrong answer and financed the right one with knowledge instead of money.

In 1953, in a small San Diego lab, the three-person Rocket Chemical Company set out to solve an unglamorous aerospace problem: rust. Missiles and aircraft skins corroded, and the team hunted a formula that would displace water — push moisture off metal — to stop corrosion before it started. Chemist Norm Larsen worked the problem the way chemists do: attempt, test, fail, adjust, again.

The lab notebook tells the story with beautiful bluntness. Water Displacement, first formula: failed. Second: failed. The count rolled on through the thirties. On the fortieth attempt, it worked — and with no marketing department to prettify things, the working formula's notebook label simply became its name: WD-40. The aerospace industry adopted it to protect the outer skin of the Atlas missile, and the product might have lived out its days as an obscure industrial supply.

Except for one telling detail: workers at the plant kept sneaking cans home. The stuff fixed squeaky hinges, freed rusted bolts, started damp engines. When your own employees are smuggling the product out the door, you have learned something no focus group can teach. In 1958, the company put WD-40 in aerosol cans and sold it to the public.

Two more details deserve to be famous. The formula was never patented — a patent would have required publishing the recipe, so the company chose to keep it a trade secret instead, and has kept it for seventy years. And the company essentially sold this one product, born on the fortieth try, for decades — into an estimated four out of five American households. The thirty-nine failures aren't hidden in a vault somewhere. They're printed on the can, in the name, on the shelf of every hardware store on earth.

By the time Edison chased the electric light, he was already famous — and already intimate with being told no. Nearly deaf since childhood, largely self-educated at his mother's kitchen table after formal schooling gave up on him early, he had built his Menlo Park, New Jersey laboratory into something new in the world: an invention factory, where failure was not an event but a process, run daily, on purpose.

The incandescent lamp problem was brutal in a specific way: the concept was obvious and the material was not. Everyone knew a filament glowing in a vacuum could make light; the filament itself burned out in minutes. Edison's team attacked the problem with industrial patience, testing material after material — metals, fibers, threads, eventually thousands of plant materials from around the world, including bamboo carbonized into fragile threads. The failures piled up in numbers that became legend.

Edison's response to those numbers is his real legacy. He insisted, in words he repeated in various forms all his life, that he hadn't failed — he'd found thousands of ways that didn't work, each one narrowing the road to the one that would. It sounds like a motivational poster until you realize he meant it operationally: every dead filament was logged, catalogued, and subtracted from the search space. Failure, systematized, became data.

In October 1879, a carbonized filament burned for more than thirteen hours — and on New Year's Eve that year, Edison lit up Menlo Park for the public, crowds arriving by special trains to see streets glowing with the impossible. The man whose schoolmaster thought him slow had, by the end, over a thousand patents and had changed what night looks like on planet Earth.

Before the castle, there was a string of collapses that would have ended most careers. As a young commercial artist in Kansas City, Disney was let go from a newspaper job — the story has passed into legend that an editor found him short on imagination and good ideas. His response was to start his own animation studio, Laugh-O-Gram, at twenty-one. It went bankrupt within two years; near the end he was reportedly living in the office and eating cold beans from the can. In 1923 he bought a one-way train ticket to Hollywood carrying a cardboard suitcase and about forty dollars.

He rebuilt — and then came the cruelest cut of the sequence. By 1928 Disney's studio had a hit character, Oswald the Lucky Rabbit. When he traveled to New York to negotiate better terms, his distributor revealed the trap: the distributor, not Disney, owned Oswald's rights — and had quietly hired away most of Disney's animators. He walked in owning a hit and walked out owning almost nothing. The legend holds that on the long train ride home, rather than compose a surrender, he started sketching a new character. The mouse.

Steamboat Willie premiered months later with synchronized sound — a first that made Mickey Mouse a phenomenon — and this time Disney owned every whisker. But he wasn't done betting. In the mid-1930s he poured everything into a project the industry openly called "Disney's Folly": a feature-length animated film, something no one believed audiences would sit through. The budget ballooned to several times the original plan; he borrowed heavily against the studio's future to finish it. Snow White and the Seven Dwarfs premiered in 1937 to a standing ovation, became the highest-grossing film of its era, and built — literally, with its profits — the Burbank studio where everything since was made.