Marketing Intelligence Service, which tracks new product launches, recorded more than 100 tests of new products each year from 1990 through 1994, but the number of tests has hovered in the 90s in recent years. And the 43 tests recorded through the first six months of 2001 established a pace for the fewest tests in more than 10 years. Such numbers only paint a partial picture of test marketing, however, because the bulk of testing covers smaller initiatives-such as new advertising and marketing approaches, media mix, and changes in price, assortment or other variables.
Valerie Skala, VP-analytic product management and development for Information Resources Inc., acknowledges that test marketing has fallen out of favor in recent years. But she also believes the scale may be tipping back toward testing.
"It tends to be somewhat cyclical," she says. "Right when we start despairing that test marketing is dead, I've just had a major client on the East Coast beg me to come out and give a presentation to their research team to rebuild their competency with in-market testing."
Ms. Skala wouldn't name names but in a presentation to analysts last month, James Kilts, new chairman-CEO of Gillette Co., ridiculed his company's analytic capability, comparing Gillette's marketing planning to that of a pilot who knows how fast but not where the plane is going.
At Procter & Gamble Co., too, President-CEO A.G. Lafley has talked recently of a need to reinstate "the good old-fashioned discipline that P&G has always been known for" as it strives to return to new-product success rates of two thirds or better. Under Mr. Lafley's predecessor, Durk Jager, P&G trimmed use of test marketing as it moved to get faster, reserving in-market tests mainly for new products that would require investment in new plant and equipment. That led to some high-profile failures, however, such as Pampers Rash Guard, an ultra-premium diaper launched in 1999 that failed to meet sales goals.
Mr. Lafley hasn't said exactly how P&G plans to reinstate old-fashioned discipline without reverting to old-fashioned product development cycles. But among approaches he mentioned in March was test-market simulation. P&G has launched ConsumerAisles.com, an online test-market simulation system.
Test marketing, however, hasn't always been a certain recipe for success. Two brands that P&G pulled the plug on last month-Fit produce rinse and Olay cosmetics-were among the most heavily tested in P&G history.
Fit began test marketing in August 1994 in Denver and found its way into additional test markets in Charlotte, N.C., Cedar Rapids, Iowa, and Midland, Texas, before rolling nationally last summer. Nationally, Fit logged $16 million in sales through May 20, according to IRI, which is $5 million less than the brand got in measured media support, according to Taylor Nelson Sofres' CMR.
"It's important to understand that Fit was a totally new product for us in a period when we had not introduced a new product category in some time," says a P&G spokeswoman. "It's not surprising for learning markets to continue for a few years, and, after all, we were helping consumers to develop a whole new habit. We were unable to replicate the marketing intensity that delivered success in test market as we expanded Fit nationally."
P&G began a three-year test of Olay cosmetics in Evansville, Ind., in 1994, then launched the product nationally in 1999 after a successful run in England. It became a $90-million-a-year business in the U.S.-significantly bigger than most new consumer products-but never reached P&G's $300 million projection.
A P&G spokeswoman says one factor was timing. By the time Olay cosmetics rolled nationally, Revlon's Almay and Johnson & Johnson's Neutrogena were also staking claims to the skin-health positioning Olay sought.
The lesson: P&G can't sacrifice speed in order to improve its success ratio. "Essentially we are expanding our tool chest of testing techniques, but it's presumptuous to assume that we are looking to do more testing," a P&G spokeswoman says. "We will still look for in-market experience through a range of techniques, which may or may not include traditional test marketing, so we can learn quickly and efficiently."