What Do You Want to Know?

Among the myriad methodologies to study a shopper -- online surveys (the most popular), virtual shopping, neuroscience -- how does a brand choose its path? Is qualitative better than quantitative? Is there a standard process; does one method lead to another?Truthfully, the answers to the questions above begin with what the company wants to learn. While researchers do favor some methods over others and debate process, they agree that it starts with what you want to know about your shopper and your product. "There's no specific formula on what gets done first or second, or any specific process," says Liz Boyd, group manager, retail insights for Anheuser-Busch, St. Louis. "What we strive to do is to focus on the business issues. Something more exploratory might lend itself to a qualitative study. 'The lift I'm going to get if I put the display here' is more quantitative."Herb Sorensen, president of TNS Sorensen, Troutdale, Ore., says, "All research needs to start qualitatively. The first thing you do as a scientist is sort and compartmentalize. The next thing is, 'How many of these do we have, and how many of those do we have?' You need numbers eventually."For ConAgra, Omaha, qualitative research is used when the company wants to know about shoppers' thoughts or emotions, and quantitative when they're getting down to brass tacks about the potential success of a specific program, says Liz Mohr, director of shopper insights, based in Naperville, Ill. "We pick the tool to match the need," she says. For example, the company might put together a summer program around Hebrew National hot dogs, Hunt's ketchup, and Gulden's mustard. Initially, it wants to know "the emotional mindset of mom" to find out what she might respond to positively, then moves to a quantitative study.Research MethodsPepsiCo, Purchase, N.Y., often starts with qualitative research, especially when still firming up the questions it needs answered, always keeping in mind that the smaller sample sizes used for focus groups or shop-alongs aren't necessarily projectable to the general population, says Sonja Mathews, director of customer strategy and insights."The thing that is very important in shopper research, in particular, is being able to covertly observe the consumer," she says. "I call it 'gentle stalking' because if you watch them, and they know you're watching them, they tend to not only change their behavior and notice things too much, but also be a little too cerebral in describing their decisions." PepsiCo runs a lab in Plano, Texas, that looks like a mock store where marketers on its various brands can observe how people shop for its products. Research that involves observing consumers is always more telling, Sorensen says, but it's also less efficient and does not provide the backstory of what's on a consumer's mind. "An observation is worth 1,000 words," he says. "Talking to people about their experiences, expecting them to be able to report those to you, can be a basis for ending up with shared misinformation.""Online research often brings people undone because it talks about a self-reported, 'I think this is what I did,' vs. what they actually did," agrees Jeff Froud, senior strategic partner with OgilvyAction, Chicago.Dreyer's Grand Ice Cream, Oakland, Calif., recently went the observational route, working with virtual reality shopping vendor Decision Insight, Kansas City, Mo. The brand expanded its offerings in single-serve ice cream cups several months ago and needed to answer whether to shelve it in a brand block or with other single-serve cups. Decision Insight created life-like sets for Kroger in Michigan and Ohio, and Von's in California, tapping 2,500 participants for feedback.The research showed that -- with the exception of health-conscious consumers who seek out brands like Skinny Cow (made by Dreyer's) and Weight Watchers -- people were twice as likely to make purchases when the single-serve cups were kept together, resulting in a 33% sales boost over brand-blocking."We got an answer that informed us in a major way," says Russ Onish, director of category development and shopper insights for Dreyer's. The company turned to virtual shopping because "we wanted to represent a selection that was more than just a few items. We wanted to present it to someone the way they would see it in a store," he says, adding that an actual store would have been the ideal answer, while using methodology like a focus group wouldn't have been nearly as accurate.Miller Zell, Atlanta, likes to use a combination of different research methods to provide "checks and balances," says John Wilkins, vice president of retail and client strategy. "Most methods on their own can give some decent direction but can lead you possibly down a false path," he says. "Focus groups are one of the biggest minefields for that. We like to use focus groups directionally, or as filters for work that we're checking, vs. front-end research."For example, Aramark, Philadelphia, worked with Miller Zell to test an on-campus c-store concept using focus groups of 18- to 23-year-olds. Intended to roll out on as many as 200 campuses, the initial store design was widely panned by those surveyed."They wanted to be treated like adults, and they found it condescending when we attempted to be hip and cool," says Jim Matthews, senior corporate vice president for business development at Miller Zell. "We completely abandoned our original approach, which was tending to be a little more youthful, a little more heavy metal, a little more in your face, to the type of materials you would expect in a Starbucks."The consulting firm particularly likes shop-alongs because "it puts the participants in a more relaxed state," says D'Anna Hawthorne, strategy director. "You're getting more authentic and genuine input, vs. in a focus group there's a more formal feel to the experience." Still, she acknowledges, "any method is somewhat contrived because you're always interacting with the people you're observing."Numbers Don't LieWhile qualitative research can certainly glean useful information about people's attitudes, Henkel Consumer Goods, Scottsdale, Ariz., focused heavily this past year on an exhaustive historical quantitative study, says Mack Hoopes, manager of shopper insights."So many times, [consumers] do something different than what they tell us," he says. "You can't tell any lies when you follow what their pocketbook says. That's it. That's what they did. And that's what pays the bills."Henkel's research divided consumers into three groups: "Shoptimizers" spend "lots of hours in a week reading ads" and use most of the coupons stores receive; "Mainstreeters" read some ads but respond more to in-store stimuli; and "Carefrees" purchase whatever they want, when they want it. (An in-depth article on this study appeared on Page 12 of the January issue.)Looking across 300 categories, Henkel found that these psychographic measures predicted consumer response to price reductions, displays, loyalty card programs and other in-store stimuli much more closely than any demographic data, Hoopes says. "Income [and] size of household have nothing to do with these buying behaviors," he says. "They transcend" demographic categories. Henkel's study used somewhere between 38,000 and 50,000 respondents recruited through IRI and Nielsen, Hoopes says.Quantitative research also tends to be the focus for Barilla America, Bannockburn, Ill., says Craig Geiger, director of category development. "It can give you insights that you can apply nationally," he says. "Qualitative is good to support the discussion or argument, especially if you're trying to form a stronger relationship or partnership with a retailer. But it's not the most compelling piece of research."Barilla performs research for all of its launches, Geiger says, from some combination of syndicated data, custom research through companies like Decision Insight, and retailer shopper card data. "It's a combination of those three main drivers," he says.Recruiting RespondentsResearch once done over the phone has shifted mostly online, representing "probably two-thirds" of all market research, Sorensen says. However, CPGs often still prefer to find their subjects in person, offering them incentives of some kind for participating.PepsiCo likes to recruit shoppers on-site for retailer-specific research, Mathews says. "By doing it this way, we assure our retail partner that we're talking to consumers who normally go into their store," she says. Online panels through third parties provide the ability to control for factors like age, gender and shopping habits, Boyd of Anheuser-Busch says. "We want 'X' number of beer shoppers between the ages of 'X' and 'Y,' and we want this gender mix," she says. "You set your specs, and you do the recruiting."Onish says Dreyer's relies on suppliers to figure out how best to find respondents and pays for their time, although the company is careful not to influence people's responses. "We don't tell them, 'We're with Haagen-Dazs, and here's $20, can you answer these questions about Haagen-Dazs vs. Ben & Jerry's?' Anything that would create bias, we don't do that."Insights Into ActionPutting data to use in-store is generally more challenging than collecting it, CPGs and consultants say. "There are all sorts of barriers; I don't think any are insurmountable," says Onish. "You need to get buy-in on the recommendation, you need to be talking to the right people, you need to represent the recommendation in a coherent manner," he says. "You need to involve other functions within your organization or the customer's. You need to get the customer's attention that this is an important enough priority."The In-Store Marketing Institute and Shopper Marketing's annual Trends Report (see last month's issue) asked CPGs which was more difficult, and nearly 60% said activating an insight. This was more of a problem for the biggest companies (those with more than $1 billion in sales).Sorensen says hundreds of millions of dollars are spent each year on P-O-P, and a lot of it gets thrown away because it doesn't get executed properly. "The barriers are knowing what to do in the store, what would really work," he says, recalling a package with a Ziploc top that had graphics on the front face but did not stand up properly on the shelf. "They laid them down" in the store, he says. "All the shopper saw was the bottom, which didn't have the graphics on it, and sales plummeted."Bringing in the retailer as a partner in garnering insights makes for a much shorter road to execution, says Wilkins of Miller Zell. His colleague Hawthorne says such collaborations have been on the rise given the state of the economy -- and even, when retailers ask, collaborations among competitors.Most CPGs say they happily partner with retailers, but few have done so with competitors. "There are pockets of it happening," says Froud of OgilvyAction. "I'm not sure how successful that will be."ConAgra relies on retailers to be honest brokers in such situations, which have come up on occasion, Mohr says. "It's about coming up with win-win solutions for both of us," she says. "We are competitors. You do need to make sure that you are remaining competitors throughout the process.""That's somewhat more rare," agrees Mike Pishvanov, associate director of shopper insights at Johnson & Johnson, Langhorne, Pa. "You want to try to keep some of the proprietary nature with the research, but we certainly share with [retailer] customers."

What Do You Want to Know?

P2PI MEMBERS

JOIN P2PI

Related Topics

Related Images

Business Topic

Company