When you hold one of OXO International's household tools in your hand, it feels right. The trademark handle—thick, black, nonslip rubber called Santoprene—nestles in your palm. The small flexible ridges (or "fins," in OXO lingo) allow for a firm grip. Whether it's a vegetable peeler, garden trowel, or cheese grater, a mundane household object is made pleasing and clever.
Before that, though, come months and years of painstaking trial and error. And you rarely hear about the experiments that don't work out. "Everybody talks about their successes, but the failures, the mistakes, are the most interesting things," says Alex Lee, OXO's president. "Our wrong assumptions lead to the best learning."
That learning has made OXO, founded in 1990 and now owned by Helen of Troy Limited, one of the most respected design shops around. Thirty of its products are part of the permanent collection at the Cooper-Hewitt, National Design Museum in New York. "By bringing design back to the basics—efficient and comfortable performance—they've elevated it," says exhibitions curator Matilda McQuaid. The payoff from that elevated design: OXO has been profitable since its first year; revenue has increased steadily, at a compound annual growth rate of more than 30% since 1991, to more than $100 million in 2004. Here are a few of the company's most memorable missteps—behind-the-scenes stumbles that ultimately led to smarter, more-intuitive products.
Mistake No. 1: All bagels are alike
After learning that a lot of Sunday morning emergency-room visits involved bagel-slicing injuries, OXO sensed an opportunity. The engineers and designers set about creating a safer slicer that would hold bagels and keep users' hands out of the way. They planned to introduce the product at the annual International Home & Housewares Show in Chicago in 1997. But when the team tried a demonstration just before the show—yikes!—the bagels slipped and slid around in the slicer. Turns out Chicago bagels are puny compared with their hefty New York counterparts used in the development and testing.
One of the company's guiding philosophies, dating back to OXO founder Sam Farber, is universal design. A product's function should be immediately apparent, and anyone should be able to use it. OXO had its nationwide sales force send in every bagel they could find. The samples helped engineers design a more accommodating bagel slicer, not simply a New York bagel slicer. They extended the internal ribs deeper into the device to hold small bagels and also made them flexible enough to contain large ones.
More recently, OXO made a similar mistake with a new toilet brush, which, as it turned out, didn't fit one-third of toilets on the market. Which raises the question: Shouldn't there be a way to anticipate problems like those posed by bagels and toilets, and cut out all the trial and error? As frustrating as the "rework" is, says Lee, it's part of the design process, a necessary by-product of experimentation. "You could design a process to catch everything, but then you're overprocessing," he says. "You kill creativity. You kill productivity. By definition, a culture like ours that drives innovation is managed chaos."
Still, some lessons do stick. When it came time to test its first mango splitter, OXO identified the most common varieties of mango sold in the United States and designed the splitter accordingly.
Mistake No. 2: You solve one problem but create another
In the late 1990s, "OXOnians," as they call themselves, noted that popular travel coffee mugs weren't spillproof, and were too cumbersome for drivers. OXO's engineers spent months creating a mug with an airtight seal and an easy-to-use push button. But their solution had a fatal flaw. "With all this steam and pressure building up in the mug, it was like a volcano waiting to erupt," says category director Abigail Levy. Worse still, because the button was on the side, you would naturally tilt the mug back, then open the spout, and the coffee volcano would erupt in your face. That mug got pulled.
For the second iteration, the team invented a small reservoir inside the top, which prevented the eruption. They also moved the button to the top, which makes users likelier to push it before lifting the mug to drink. The LiquiSeal Travel Mug is now one of OXO's best-selling items (out of more than 750). "We're not afraid of a slow build," says Levy. "We're designing products to last a long time."
Mistake No. 3: Solving problems that people don't want solved
In Scandinavia, the wire cheese slicer is a household staple. To OXO, it had an obvious flaw. Consumers use the slicers daily, and the wire eventually breaks and needs to be replaced. That's the sort of work-around that gets OXO's attention.
The company's remedy was to incorporate a blade instead. But when OXO tested the product in Denmark, customers weren't interested. "The wire slicer is the tradition," says category director Michelle Sohn. "Even though it has a problem built into it, they kind of like the problem." So instead, OXO is working on a more durable wire.
Mistake No. 4: Thinking you're done
OXO's customers call or write all the time, suggesting improvements or new products. Most, says Lee, are obvious or impractical ideas. But sometimes OXO gets a gem. One woman wrote in asking why the handle on the potato masher was vertical instead of horizontal. "The lightbulb went off right away," says Lee.
OXO introduced a new masher in 1999 with, among other things, a horizontal handle that enables users to push down with more force. If Lee is disappointed that OXO didn't come up with the idea on its own, he doesn't let on. He's more enamored of the design itself, its utter simplicity, and, in hindsight, how obvious it was. The new masher is a reminder never to stop questioning the norm and looking for improvements. Being content with a design is one mistake that Lee and his fellow OXOnians try to avoid at all costs.
Chuck Salter (firstname.lastname@example.org) is a Fast Company senior writer based in Chicago.
A version of this article appeared in the October 2005 issue of Fast Company magazine.