Smells Like Brand Spirit

Dr. Eric Spangenberg, dean of the college of business and economics at Washington State University, ran a test in a clothing store in the Pacific Northwest to determine how scent affected customers by gender . He diffused a subtle smell of vanilla in the women's department and rose maroc (a spicy, honeylike fragrance that had tested well with guys) in the men's. The results were astonishing. When he examined the cash-register tapes, he found that receipts almost doubled on the days when scent was used. However, if he reversed the scents, diffusing vanilla with the men, rose maroc with the women, customers spent less than average. "You can't just use a pleasant scent and expect it to work," he says. "It has to be congruent." Similarly, he says, the fragrance has to make sense with the product or environment it's supposed to enhance: "When you go into Starbucks, you don't expect to smell lemon-scented Pledge."

Alex Moskvin, director of BrandEmotions at International Flavors & Fragrances Inc. (IFF), spends a lot of time tinkering with variables, attempting to match the right scent to the right brand. He begins by developing a demographic profile of a target customer and by identifying the elements of the brand that consumers find most appealing. Moskvin says he studies the DNA of the brand and its relationship to consumers -- even those who have become disaffected -- to figure out what resonates olfactorily. "It's important for companies like ours to understand the emotional communication of the fragrance and to have a point of view on that," he says. In designing a hotel fragrance, for example, Moskvin would want to know if the chain was positioned as a family-friendly hostelry (think chocolate-chip cookies) or an haute couture, Zen-like retreat (think sandalwood or hinoki). "We want to capture a smell that makes people feel part of the club," he says.

As many global marketers have learned, however, the scent of the club varies depending on where the clubhouse is located. At a seminar in Moscow, for example, Lindstroem polled the audience on their favorite aroma. "Freshly washed clothes, hanging on a line in subzero weather," they responded enthusiastically. Hispanics, experts say, are partial to vanilla. The Japanese like apple.

And to add to the complexity, much as in the fashion industry, fragrances follow trends. Indeed, says Ted Kesten, CEO of the international fragrance house Belmay, food-related scents -- chocolate, vanilla, pecan pie -- are the new black this year, at least in the United States. But that may not hold true in Barcelona or Bangkok. Five years ago, for example, fruit scents were popular in the States, but not in Europe. "Global brands are very tricky," he says. "When you're launching an international scent, you try to stay away from the love-hate extremes and find something in the middle with wide acceptance."

In 1994, Dr. Hirsch gave a paper on the effect of scent on consumer spending to a group at the Bally Hotel in Las Vegas. At the end of his talk, an audience member stood up and challenged him. "Hirsch," he said, "the odors aren't affecting consumers; they're affecting the salesmen, who are being more friendly." Hirsch was stumped. It was a variable he hadn't considered.

Walking back to his room, he passed a bank of slot machines, and three cherries rang up in his brain: What if he designed an experiment using aroma and one-armed bandits -- a totally salesperson-free control group!

So he set up a scent-free control area, and another with a faint, pleasant scent. Then he stood back and let the low rollers ante up. Afterward, he compared the test sample's receipts with the weekend before the study and the weekend after. In the control area, there was about a 3% uptick in bets, defining the study's margin of error. But in the scented area, receipts were up 45%. When the study was replicated using a higher level of odorant, there was a 53% increase. The study was reported in the journal Psychology & Marketing. Since then, Hirsch says, most casinos use some form of scent, although all will deny it.

Despite instant comparisons to the famous subliminal advertising ploys first discussed by Vance Packard in The Hidden Persuaders, Hirsch claims to be unconcerned by the ethical implications of this sensory manipulation. For one thing, he says, the scents have to be used in high enough concentrations to be noticeable, so they can't be secret. "When you use lower odors, it's less effective than when you use higher odors," he says. "So why would people bother using subliminal smells when higher odors have greater efficacy?" (Other researchers have warned that cranking up a scent too high, however, can be a turn-off.)