5,127 prototypes: James Dyson and the vacuum cleaner nobody wanted to make

In the late 1970s, James Dyson was annoyed with his vacuum cleaner. As the dust bag filled, its tiny pores clogged and the machine lost suction. It is the kind of everyday irritation most of us shrug off. Dyson did not.

The idea for a fix came from somewhere unexpected: a sawmill. The mill used a cyclone — a spinning column of air — to pull sawdust out of the air without a filter that could clog. If it worked for sawdust, Dyson reasoned, why not for household dust?

Five years and 5,127 attempts

What followed was not a eureka moment but a grind. Dyson built about 5,127 prototypes over five years, changing one thing at a time and testing again. He filed his patent for the dual cyclone in 1980.

He already knew something about invention. His first product, the Ballbarrow — a wheelbarrow with a ball instead of a wheel — had been featured on the BBC's Tomorrow's World. But a better wheelbarrow is a small market. Vacuum cleaners were in almost every home.

No one would make it

With a working design, Dyson went to the established manufacturers. They turned him down. One reason, in an industry that made good money selling replacement dust bags, was that a bagless machine threatened part of their business. No manufacturer or distributor would take the product in the UK.

So he went abroad. In 1983 a version called the G-Force went on sale in Japan through catalogues, at a price equivalent to about $2,000. It won the 1991 International Design Fair prize in Japan. The licence income kept the project alive.

Doing it himself

In 1991, Dyson set up his own company in Malmesbury, Wiltshire. Its research centre and factory opened in 1993, and the first Dyson-branded vacuum, the DC01, followed. Within 18 months, the DC01 had become the best-selling vacuum cleaner in the UK.

The idea had to be defended as well as sold. In 1999 Dyson sued Hoover for copying his design. The High Court found that Hoover had copied a fundamental part of his patented design, and Hoover agreed to pay £4 million in damages.

The company kept growing — and kept making choices that drew attention. It moved vacuum cleaner production to Malaysia in 2002, and moved its headquarters to Singapore in 2019.

Why it took so long

It is tempting to read the story as a straight line from idea to success. It was not. For years there was no income from the vacuum at all, and every prototype cost time and money. The Japanese licence mattered because it paid for the next stage. Without it, the project could easily have stopped at prototype number two thousand.

That is true of most new products. The first version that works is rarely the version that sells. Between them sit months of small changes: a cheaper part, a simpler shape, a way to make it in a factory rather than a workshop. The slow, unglamorous middle is where many good ideas die — not because they were wrong, but because the money or the patience ran out first.

What a small business can take from it

Start with a real annoyance. The best product ideas often come from a problem you have every week. Dyson did not set out to invent a vacuum — he set out to stop his own one losing suction.

Borrow from other industries. The cyclone was not new. It was already working in a sawmill. Many good ideas are an existing solution moved to a new place.

Expect the established players to say no. If your idea threatens part of how they make money, they have a reason to turn it down that has nothing to do with whether it is good.

Find another door. When UK manufacturers refused, Dyson licensed abroad and used that income to fund the next step. For a small business, the equivalent may be a smaller market, a different sales channel, or selling to businesses before consumers.

Protect what makes you different. The patent turned a design into something that could earn money and be defended in court.

Count the attempts, not the failures. 5,127 prototypes means 5,126 versions that were not right yet. Each one taught him something. Most businesses give up after far fewer tries.

The lesson is not that everyone should spend five years in a shed. It is that persistence works best when it is organised: change one thing, test it, write down what happened, and go again.