How P&G Conquered CarpetSarah Lacy
Slide Show >>
It was November 19, 2003, and Bob Godfroid was in unfamiliar territory. He was crouched on the floor of a small nondescript storefront in Palo Alto, Calif. The windows were frosted over so no one could see in, and scraps of very messy, multicolored carpets were scattered throughout the room. About a dozen engineers and product designers were hunched over them frantically -- cleaning. One was fiddling with a scary looking suction gun, while a whole team was rolling balls of Play-Doh around.
After nearly 12 hours, Godfroid, a chemist who had worked at Procter & Gamble (PG) for 10 years, drove back to his hotel scared. He had faced plenty of innovation challenges at P&G, like improving Tide's ability to lift grass stains off jeans, but now he was charged with the nebulous task of finding a way to "Swiffer" a carpet.
"You should have seen the condition of that room," he says. "It looked like a bomb went off." Driving back to the hotel, fear set in as he thought about what he would say to his boss when he flew back to Ohio in two days. "I don't know if we're going to come out of here with anything other than a bunch of pictures of a trashed room," he thought to himself.
On that scary day, the 20 month development of P&G's new CarpetFlick began. It all started with a simple observation: P&G's products clean more hardwood, tile, and linoleum floors than anyone's, but 75% of U.S. floors are carpeted. P&G didn't want to get into the vacuum business, but what if it could find a way to "Swiffer" a carpeted floor?
The Swiffer had been one of P&G's greatest design triumphs of recent years. Released in 1999, it's a very simple sweeper with a rectangular head and a long pole that swivels around, making it easy to get into nooks and crannies. Disposable cloths fit on to the head and trap hair and dust, instead of just pushing it around like a broom.
According to P&G, Swiffer has a 75% share of the quick-clean market, which would put revenues around $750 million a year. It's one of the most profitable products in P&G's arsenal and the anchor of a product line that now includes a mop-like version, a duster, and more.
"It has been a pop-culture phenomenon," says Karl Ronn, vice-president for research and development for P&G's home-care division. "It was on the cover of Rolling Stone. If you would have bet me that one of P&G's brands would have been on the cover of Rolling Stone with Jessica Simpson, I would have lost that bet."
But the Swiffer's success simply extended P&G's market share in cleaning hardwood floors. The company was still missing a huge opportunity in carpets. P&G contracted with Palo Alto (Calif.)-based design consultancy Ideo, which had done a few one-off projects for P&G in the past, like redesigning a toothpaste tube.
About four years ago, P&G and Ideo started a more creative and collaborative arrangement, in which they would work together to invent new products, not just improve on existing ones. Out of that collaboration came Pringles Prints, potato chips with trivia facts printed on them, and Mr. Clean Magic Reach, a wand with a disposable cleaning pad that allows people to clean most of their bathrooms without getting down on their on hands and knees.
But CarpetFlick was a new product going after a totally new market. It was ambitious, and Godfroid would face challenges every step of the way. As he learned that first day in the Ideo Lab, this was going to be a product-development cycle unlike anything he had ever experienced.
The first step was research. Godfroid and the Ideo team, led by Mike Strasser at that point, went into people's homes snapping photos and asking questions about how they cleaned their carpets.
There was a young mother who complained that the noise of the vacuum scared her child, but she had time to vacuum only when he was asleep. There was an older woman with a busted knee who relied on two vacuums -- a heavy one for once a week cleaning when she took painkillers for her knee, and one she could easily lift for spot cleaning.
Carpet sweepers weren't much better. They were hard to push, and hair and lint would get tangled in the brushes. And often the brushes would miss something like a paper dot. "These people were crying out for better solutions," Godfroid says.
Even more convinced of the need for a carpet-worthy Swiffer and armed with some inklings of how it needed to differ from a vacuum, Godfroid and the Ideo team started what they call a "deep dive." During the two day workshop in November, 2003, Ideo designers spent hundreds of dollars at the local Ace Hardware store for all sorts of random squeegee stampers, and adhesives. About 15 designers and engineers set upon dirtied carpet squares laid out all over the room, sucking, scraping, stamping, sticking, and trying anything else they could come up with to clean carpet.
On the morning of day two, the dubious Godfroid was sitting with a few other designers around a carpet square littered with dirt and confetti. Holding a squeegee blade in one hand, he tried scraping it over the carpet. The particles sprung up in the air. He tried it again, this time with the edge of the squeegee angled more toward the carpet. Just like a child's game of TiddyWinks, the particles flew up even higher.
The team got excited as they realized that finally, they were making the dirt move. When someone grabbed a balloon and held it behind the squeegee, the static electricity attracted the popping dirt and confetti. Godfroid was starting to feel a little better about going back to Ohio.
He and Strasser scrambled to make a crude prototype they called the "Shagilator" by the workshop's 5 p.m. deadline. It was little more than a rectangular box with a rounded bottom, made out of foam board, hot-glued together, with a long, narrow slit cut at the base. The edges of the slit each worked like the blade of the squeegee, flicking dirt and crumbs up into the container, leaving the carpet clean with just one pass.
Minutes before 5 p.m., Godfroid pressed together the just-glued Shagilator, eagerly anticipating presenting it to the rest of the design team and his meeting with VP Ronn the next day. He knew this was it.
Back in Ohio, Ronn found the prototype pretty cool. But it was the middle of the fiscal year, and funds were running too low to continue the project. He could apply for more money after the first of the year, but he was worried the team would lose momentum. So he borrowed the Shagilator and marched into Gilbert Cloyd's office in the executive wing.
As P&G's chief technology officer, Cloyd had a slush fund for innovative projects. "Gil, I've got something to show you," Ronn said, dumping crushed Froot Loops onto the carpet and sweeping them up with the Shagilator. Ronn pointed out that, as it was, P&G didn't make a dime in carpets and said if Gil gave him a check, he would have a product on the market by July, 2005, some 18 months away. He left Cloyd's office with the Shagilator and a check for several hundred thousand dollars to keep development going.
Back in Palo Alto, Ideo staffers were giddy. They were cutting slits in everything they could find, and sweeping up carpets. They cut a slit in a piece of Tupperware. They cut a slit in a McDonald's Cherry Pie envelope. They cut a slit in a Pringles can, dumped out the chips, crushed them, and made the can "eat" them back up.
"We were just giggling more and more, because anything could do it," recalls Sam Truslow, Ideo's CarpetFlick project leader. "God, [we thought,] this thing is just magic."
Truslow and Godfroid took these crude prototypes back to the people they had observed at the beginning of the project. The team got some skeptical looks when they asked the sample group to, say, run the top of a butter dish with a slit cut in it over their messy floors. But it worked. "We got a lot of, 'We don't want to give this back,'" Godfroid remembers.
By early 2004, the basic concept was clear, and the team had to turn its crude foamboard prototype into something "Swifferesque." Over a six-month period, dozens of prototypes followed. Finally, the team settled on a squarish head with the now very familiar slit in the middle and a removable adhesive strip inside to trap the upflicked particles. It had a pole like the Swiffer but was ergonomically designed to go only backwards and forwards, not swivel in every direction like the Swiffer.
There were a few hiccups -- like an ill-fated move to add a hinge, so the disposable strip could "eject" out. The engineers and designers were enamored of the idea and obsessed with the challenge of making it work, but after a few weeks they conceded that it wasn't intuitive or easy and only added more cost into the manufacturing.
Still, by September 29, 2004, the design phase was done, and the CarpetFlick was being tested in 350 homes around the world. Truslow, Godfroid, and the rest of the team were pleased with themselves. About 50 top P&G executives -- including Chief Executive A.G. Lafley -- came to the Ideo offices for a meeting, and got a demo of the new product.
Truslow and Godfroid were on top of the world, demonstrating their CarpetFlick to handshakes and pats on the back, when Home Care President Jorge Mesquita pulled them into a conference room and closed the door. A problem remained -- lint and hair. The team knew the CarpetFlick wouldn't pick them up, but in early visits to demo the device in homes, people were so wowed at how the Shagilator removed other dirt they didn't care. So, following Ideo's mantra of making things as simple as possible, the team leaders decided they would ignore the hair and lint problem, focusing on tiny particles instead.
But now, as the researcher in charge of testing told Godfroid, "That hair you made the conscious decision not to pick up has come back to bite you in the ass." Ronn was a bit more politic, telling the two, "Fellas, we're still launching in August, but we're not launching this. Come down to the reception and have a glass of wine, then fix it."
They didn't even wait that long. Truslow and Godfroid went downstairs, grabbed a few bottles of wine from the bartender, and walked across the street to the lab where they sat until 2 a.m. thinking about how their elegant design could be altered to remove hair, fuzz, and lint.
The next morning, after another trip to Ace Hardware, Truslow and a dozen Ideo staffers started experimenting with sandpaper, lint rollers, Brillo pads, and more. The frenetic brainstorming started again. The team quickly decided to take the velour found on lint brushes and line the slits with it so it could pick up fizz and lint. They thought to add glue to the underside of the replaceable strip, but worried it would just stick to the carpet.
Meanwhile, in the P&G R&D labs, two engineers were also working to solve the problem. One grabbed a chopstick and glued it down the middle of the sticky paper, the side that would glide along the floor. It worked -- the chopstick kept the paper up high enough to avoid sticking to the carpet but could trap any hair or fuzz the edge of the slit couldn't flick up.
NO TIME TO REST.
As soon as Godfroid heard the news, he grabbed the phone and called Truslow. If they hustled to refine the idea, the team could still make the deadline. There was just one last tweak: The color would be changed from the familiar blue-green Swiffer hue to a bright day-glo orange. The idea was to make the product look clean and hip. And P&G wanted to impress on consumers that this Swiffer-esque device was for a totally different kind of surface.
Weeks later, new testing showed happy customers, production began, and P&G shipped the first CarpetFlicks to Europe in the last days of July, meeting the deadline Ronn had made to Cloyd back in 2003.
Truslow and Godfroid have reason to celebrate, but if the myriad Swiffer knock-offs are any guide, they'll have to get back to work pretty soon to stay ahead of competitors. Which means, no doubt, another trip to Ace Hardware.