A 55-year-old white female executive in New York City.
A 41-year-old Latino male lawyer in Cincinnati.
A 23-year-old Asian-American female grad student in Palo Alto.
A 67-year-old black male Naval retiree in Fort Lauderdale.
Conventional marketing would address all of these people differently. Conventional marketing is dead wrong.
That’s because conventional marketing breaks people into groups based on socio-economics—age, class, race, geography. These factors might be definitional for a given individual, but when we lump people together using such broad definitions, things fall apart fast.
Maybe that grad student and the lawyer are both new to wine. They’re smart and experimental, and while they have little money to spend, they want to spend it on cool wine experiences.
Meanwhile, that executive and the Navy retiree are long-time wine drinkers who enjoy throwing wine dinners for people they love. They’re also both staunch traditionalists, and value family and friendship above everything else.
“Demographics aren’t really as explanatory as they were in the past,” said Stephen Rappaport, a digital marketing strategist and corporate adviser to global brands, speaking recently at Nomacorc’s Exchange forum for wine marketers. That’s why newer marketing research approaches try to dig deeper, moving beyond the superficial demographics to identify attitudes, lifestyles, values.
For the past seven years, Rappaport has collaborated with Dr. Howard Moskowitz, an experimental psychologist, to develop new strategies for exploring what people think, then identifying how these beliefs influence their behavior. The pair dubs their approach Mind Genomics, and they’ve applied it to a range of sectors, from Law to Philanthropy, Education, Conflict, Wellness, Automotive, and—relevant to our discussion here, Food and Drink.
Their method is simple, but it probes beyond the surface attributes of a wine—the amount of tannin, flavors of cherry or apple—to include emotional cues and perceived benefits, too. “We enjoy wine within a context,” said Moskowitz, speaking alongside his colleague. That means it’s essential to look at a consumer’s response to the entire wine experience, not just to the elements found in the glass.
Constellation Brands, one of the world’s largest premium wine producers, with revenues of nearly $4 billion, has also taken a deep dive into consumer data to mine it for attitudes and beliefs. Evaluating wines’ sensory qualities and merging these with consumer preference information yields insights to inform wine product development.
“Sensory segmentation is the new competitive weapon in driving consumer preference,” said Chris Fehrnstrom, until recently Constellation’s chief marketing officer, also speaking at the Nomacorc forum. “Connecting attitudinal segments to occasions and sensory factors helps brands differentiate in a crowded marketplace.”
What’s at stake isn’t merely a matter of marketing success, it’s a matter of ethics. Rather than make inferences based superficial appearances like sex, race, and ethnicity, corporations can make inferences based on consumer-reported preferences. The results are more nuanced, more humane, more surprising—and more fun. “The reason that it’s fun is that we’re discovering things about people,” said Rappaport, smiling.
Let’s dive into both of these models, then see what together they might suggest for wine marketing strategy.
Mind Genomics by Moskowitz, Rappaport, et al.
The Mind Genomics method involves showing consumers a series of statements about a wine, then asking them whether they’d view that wine as appealing. They’ve found, for example, that “A light bodied white wine with a fresh fruity flavor” and “Made in the tradition of the greatest wine producers all over the world” received more favorable responses than “Bubbly fresh and sweet with a hint of lemony tart flavor” and “Tastes great without the taste of alcohol.”
Regression analysis of these data suggests there are definable groups of people who will respond similarly to wine attributes. Importantly, these groups transcend age and income factors.
Although the results of this study have not yet been published, Moskowitz has made draft papers available for download in this Dropbox Folder (I recommend wine professionals start with the document It! Wine Short). Their results suggest these four types of wine consumers:
This group places a strong emphasis on the sensory aspects of both red and white wines, responding most strongly to keywords about tastes and aromas. “These consumers are adventurous in their search for unique styles of wine,” according to the draft report. Elaborates “like their wines to have a wide range of tastes and sensations” and are “willing to flout traditional wines in search of new flavor experiences.” This segment actually scores low in their overall interest in wine, but they seem to be especially susceptible to discussion of a product’s attributes.
Classics love tradition—they’re interested in red and white wines made in historic regions according to traditional methods. “Any wine that is different from that detracts from overall acceptability,” according to the authors. This group also exhibits “high negative utility values for the fizzy and sweet wines” (meaning, presumably, the newer bubbly wine products—not traditional demi-sec Champagne).
These wine drinkers love the romance of wine. They connect most strongly with wine’s experiential benefits, and view wine as celebratory and engaging. “Almost anything that adds to the drinking experience is a positive element,” the authors suggest. “This could include the history of the vineyard or region from which the wine was made, or even the types of food that match the wine.” Like the Elaborates, they, too, have a low overall interest in wine, but their connections to it are the most emotional and they also respond strongly to descriptions of product features.
This group is the opposite of the Imaginers and Elaborates. They’re interested only “in a simple wine that will not cause them any trouble and is easy to drink.” They’re not persuaded by elaborate descriptions—in fact, these can decrease a wine’s perceived appeal. This preference group is weaker than the others, though, correlating with only 11 percent of red wine respondents. But it’s important to note that with some consumers, marketers’ predilection for flowery language might work against their cause.
Wine Genome Project by Constellation Brands
In 2005, Constellation Brands released the first Wine Genome Project report, a segmentation model based on original consumer research. They subsequently added to the model, incorporating analysis based on Nielsen Homescan data. Constellation recently updated the data based on a fresh survey of 4,000 Americans and nearly 3,000 Canadians. The refreshed and refactored model suggests six types of wine consumers, sketched below:
Ten percent of surveyed consumers describe themselves as wine lovers: “I love everything about the wine experience—researching purchases, reading reviews, shopping, discussing, drinking, and sharing with others,” according to the report archetypes. An equal mix of women and men, these people are generally in mid-life with a higher than average household income. Forty percent of their alcohol consumption is wine.
30 bottles on hand, 12 glasses per month, $13 average bottle price
50 average age, 50% male, $86K household income
Twelve percent of respondents are in discovery mode: “I’m new to an intimidating category. Wine is a big part of the socializing I do. I’m interested in learning more.” They’re young and mostly male, curious, exploratory. Although they’re intrigued by wine, wine constitutes only a quarter of their alcohol consumption—beer, spirits, sake, and cider also catch their eye.
7 bottles on hand, 7 glasses per month, $13 average bottle price
36 average age, 62% male, $73K household income
Eighteen percent of wine drinkers seek status above all: “How others perceive me is important. I want to live a life that impresses others. I want to make sure the wine I choose says the right thing about me.” They skew male, and have the highest household income of these six segments, although the average price they pay is on par with that of Enthusiasts and Newcomers. About a quarter of their alcohol consumption is wine.
10 bottles on hand, 7 glasses per month, $12 average bottle price
41 average age, 63% male, $91K household income
Twenty percent of consumers like wine but don’t venture far from their preferred brands: “Wine drinking is part of my regular routine. When I find a brand I like, I stick with it.” Almost half of their alcohol consumption is wine, and they drink more wine overall than any other segment. Slightly more women identify as this type, and they skew older, too.
10 bottles on hand, 13 glasses per month, $10 average bottle price
58 average age, 52% female, $70K household income
Twenty-one percent of those surveyed buy by price: “I believe you can buy good wine without spending a lot, so price is a top consideration.” They drink almost as much wine as the Everyday Loyals, but pay 20 percent less and constitute a less profitable cohort. They’re mostly female and older, and—unsurprisingly—have the lowest household income of Constellation’s tiers.
6 bottles on hand, 10 glasses per month, $8 average price
56 average age, 59% female, $63K household income
Nineteen percent of consumers are simply overwhelmed by wine choices: “I drink wine, but it doesn’t play an important role in my life. I don’t enjoy shopping for it, and find it complex and overwhelming.” This group spends less money on wine, keeps very little around the house, and doesn’t constitute a very profitable segment, yet 34 percent of their alcohol consumption is wine. Two-thirds are women.
3 bottles on hand, 7 glasses per month, $9 average price
49 average age, 67% female, $64K household income
Putting it All Together
“There’s no company that knows, on demand, what to say in every context,” Moskowitz says. Research helps. There’s no substitute for a company actually asking customers what they want, then using that information to ensure products fit their needs. He says it’s possible to get actionable information by testing even a small number of subjects.
On the other hand, it’s unwise to rely exclusively on cohorts that coalesce from qualitative human subject research. Self-reported information is notoriously aspirational, and it’s important for a company to consider quantitative—preferably outcomes-related—data, too, and to keep testing to refine assumptions.
A comparison of Constellation and Moskowitz models proves these points, because they don’t map perfectly to one another. Constellation’s model does rely on a larger corpus, but digging in, I did find overlap with Moskowitz:
- Elaborates are somewhat like Enthusiasts. They like wine’s sensory aspects. They’re adventuresome and experimental. They appreciate a range of flavors, textures, and experiences, and are happy to hear descriptions of wine.
- Classics are somewhat like Image Seekers. They value prestige, and classical wines from traditional wine regions—and the images these convey. True, some Image Seekers might wish to be viewed as contemporary or novelty-seeking, but their rejection of fizzy, sweet wines suggests a preference for Bordeaux over Barefoot.
- Imaginers share qualities with both Engaged Newcomers and Enthusiasts. They’re intrigued by wine, maybe new to it, and experimental. They like to explore wine through experiential and emotional lenses.
- No Frills share qualities with both Price Driven and Overwhelmeds, and maybe Everyday Loyals. These people are practical, thrifty, and time-stressed. Wine isn’t the most important part of their social life.
What should we conclude from looking at these two example studies? Maybe simply that people are messy and complicated. That some of them will like your wine, but not all of them will like your message. That you can’t please everybody, but you can please somebody. That you should do your research, test your hypotheses, toss out your assumptions—and never stop talking with real people.
 Both Constellation and Moskowitz, et al., reference genotype, but the term is misleading and unfortunate. We’re not talking about genetics, here, or trying to establish baselines accruing to biology. We’re talking about behavior, so the right word is phenotype. This term describes how beings interact with their environment, which includes social cues (and we all know wine is social). But let’s skip over the confusing jargon and simply call these what they are: segmentation models.