The power of brands, conscious and unconscious

Imagine yourself in your local supermarket, doing your usual grocery run. In aisle after aisle, countless brands vie for your dollars and attention. Colgate, Crest and Aquafresh toothpaste. Smucker’s, Welch’s and Dickinson’s strawberry jam. Coke, Pepsi and RC Cola soft drinks.

Like most consumers, you probably prefer one over another in dozens of categories that are collectively called consumer packaged goods. This class of goods includes things with a relatively short life cycle such as food, drinks, cosmetics and cleaning products. These items are used and replaced quickly, compared with durable goods like appliances and cars. True product innovations in this realm are rare, and actual distinctions between brands are typically slim. Yet reaching for your go-tos is almost automatic, and economists have long tried to determine why.

It’s especially curious because, as research has repeatedly shown, consumers in blind taste tests routinely fail to pick out their preferred brands. And they will happily purchase name-brand products such as Advil even when identical generic products sit right next to them on the shelf for a fraction of the cost.

People’s brand preferences are deep-seated and long-lasting. It’s not as simple as one brand being most popular across the board, though. Different brands dominate market shares in different geographic regions. In some cases, these market shares have generally been considered to remain stable over remarkably long periods of time, including those of Coca-Cola, Wrigley chewing gum and Gillette razors. Among food and beverage brands, nearly 50 percent of those that were dominant in the US in 1923 were still among the top five in 1997.

Precisely why people form lasting brand attachments to more or less  indistinguishable products is a challenging question, says Stanford marketing professor Bart Bronnenberg. Psychology, geography, childhood experiences and advertising all exert some influences, concluded a 2017 paper in the Annual Review of Economics by Bronnenberg and Jean-Pierre Dubé, a marketing professor at the University of Chicago Booth School of Business. But, they note, finding definitive answers is made harder by the fact that long-term, multi-decade purchasing data for individuals and households are difficult or impossible to obtain.

“We’re just scratching the surface here,” Bronnenberg says. “I think the exact mechanism through which this stickiness comes about is still unknown, but people are researching it, and we have several good answers.”

And while researchers have gained insight into how brand loyalties are formed and persist in the brick-and-mortar marketplace, the shopping landscape is changing rapidly. How consumer buying habits may change — and brand loyalties along with them — in the age of online shopping may pose even trickier questions.

Here are some of the best explanations researchers have for why you reach again and again for Crest, Smucker’s or Tide. And how those habits may change as shopping gets smarter and more connected.

“Consuming” the brand

Brands were originally used as symbols of quality. They communicated to consumers that this cow was raised by a reputable farmer, that this beer was brewed with untainted water, or that this loaf of bread was free from sawdust and other fillers.

“A century ago, as branding was really emerging, pundits tended to think of the brand as a means to identify the supplier of a commodity,” Dubé says. “But we have subsequently learned that consumers derive other benefits from the brand.”

In a classic study conducted in 1964, hundreds of participants were asked to rate different kinds of beer. With labels on the beers, the participants rated their go-to brand higher than the others. But when the brews were unlabeled, participants showed no preference. In other words, the brand alone made the beer seem to taste better. “It’s as if the brand is a complement to the product itself,” Bronnenberg says. “It adds value, or adds to the enjoyment.”

This added value can extend from psychological effects to physiological ones, too. For instance, in a 2013 study, the antihistamine Claritin proved more effective at relieving symptoms in participants who took the drug and were then shown a Claritin advertisement, versus those who took Claritin and were shown a Zyrtec commercial. Here, Dubé says, is evidence that “buying the brand makes me ‘feel better.’ I literally consume the brand itself.”

The “cognitive miser”

In other cases, brand selection may merely be a matter of convenience. When you’re in the grocery store searching for all of the items on your shopping list, the kids are begging for lollipops and you’re trying to plan dinner ideas for the upcoming week, how much mental bandwidth do you have to spare?

People are, as psychologists say, “cognitive misers.” We tend to think and solve problems in the simplest possible ways. To help make choices, we form heuristics, or mental shortcuts.

“In the grocery store, you’re making choices for sometimes 40 or 50 categories, each of which have like 40 or 50 alternatives,” Bronnenberg says. “It’s impossible to do this in a timely fashion unless you have these shortcuts like brands.” When it comes to choosing toilet paper or bottled water, the quickest way to make a decision might simply be to buy the brand you recognize, or the one you purchased last time. Here, Dubé says, the brand acts as a beacon, steering you toward things you’ve bought before.

And one snap decision may well turn into a behavioral pattern. As long as the product doesn’t let you down, that’s probably enough to earn your loyalty. Or, more accurately, you can’t be bothered to not be loyal. “You tried a brand and it worked,” Dubé says. “Why waste time finding something else?”

Home is where the brand is

Another possibility, raised in a 2009 study by Bronnenberg, Dubé and Sanjay Dhar, is that brand preference can stem from where you live. The team studied which brands dominate market shares in different regions, tracing each brand’s city of origin and the timing of its entry into other markets.

Bronnenberg and colleagues found that the dominant brand in a given region often reflects which brand hit the shelves there first, even if that occurred more than a century ago. In fact, a brand’s market share in its city of origin is approximately 20 percent higher than in a city 2,500 miles away.

And while this advantage erodes with distance from the city of origin, it does so only gradually. For example, Portlanders prefer Folgers coffee, first sold just down the coast in San Francisco in 1872, while Clevelanders prefer Maxwell House, which launched in nearby Nashville in 1892. This pattern held true across dozens of different categories of consumer packaged goods, from ketchup and mustard to bagels and breakfast sausages.

Leave a Reply

Your email address will not be published. Required fields are marked *