How Many Reasons Do You Need to Like BMWs?

If you were asked to come up with one reason to drive a BMW, what would it be? What if you were asked for ten reasons? Now imagine your opinion of the BMW after searching your mind for reasons to drive it; would your evaluation be more favorable after coming up with one reason or ten? Contrary to the intuition that “more is better”—and thus that a consumer’s favor will increase as the number of reasons to drive a BMW increases— a recent study found that consumers like a BMW car more when they were asked to think of one rather than ten reasons to drive it. This outcome is thought to occur because people base their judgments on how easy or difficult it is to think of the reasons—or “retrieval ease”—rather than on the content or number of reasons they generated. Thinking of one reason to drive a BMW is easier than thinking of ten, so people evaluate the BMW more favorably when asked to think of one reason.

Does this mean that the “more is better” theory should be replaced by “less is more”? Perhaps a better question to ask is, “When will consumers base their evaluations on retrieval ease and when will they base them on the content of the information they consider?” Alice Tybout and Brian Sternthal (both of the Kellogg School of Management) and their colleagues—three graduates from Kellogg’s doctoral program in marketing: Prashant Malaviya (INSEAD), Georgios Bakamitsos (Tulane University), and Se-Bum Park (Korea Advanced Institute of Science and Technology)—have set out to answer the latter question.

These researchers hypothesized that consumers’ access to information about a brand determines how they form their judgments. When consumers have limited brand knowledge, they recognize that thinking of reasons to purchase that brand will be difficult and thus the ease they actually experience is not informative. Similarly, when consumers are highly knowledgeable about a brand, they realize that thinking of reasons to purchase it will be easy and thus uninformative. Both kinds of consumers rely on the content of the information (the reasons) they consider, and their evaluations are more favorable when more reasons to purchase the brand are brought to mind. However, when consumers have moderate brand knowledge, they lack clear expectations about how easy or hard it may be to think of reasons to purchase the brand. Consequently, the ease with which they can think of reasons forms part of their judgment about how much they like a brand. Thus, one reason results in a more favorable evaluation than ten because it is easier to bring one reason to mind.

Tybout and colleagues conducted several studies to test their hypothesis. In an initial experiment, participants (who were known to have limited knowledge about Hyundai and moderate knowledge about BMW cars) were assigned to one of four conditions: they were asked to think of either one or ten reasons to drive a Hyundai or to think of either one or ten reasons to drive a BMW. As expected, when asked to think of reasons to drive a car about which they had little knowledge (Hyundai), participants were more favorable when they were asked to think of ten reasons rather than one. This outcome implies that content served as a basis for their evaluations. By contrast, when asked to think of reasons to drive a car about which they had moderate knowledge (BMW), participants were more favorable when they were asked to think of one rather than ten reasons. This finding suggests that these individuals reflected on how easy it was to retrieve reasons as the basis for their evaluations.

Additional support for the hypothesis was found in a second study, which used a similar method. This study was conducted in Korea with participants who had little knowledge of Saab but moderate knowledge of both BMW and Hyundai. In this context, evaluations of Saab, just as with Hyundai in the Tybout experiment, were more favorable when participants were asked to think of ten reasons rather than one. The evaluations of both BMW and Hyundai (brands about which the participants were moderately knowledgeable) were more favorable when participants were asked to think of one rather than ten reasons.

A third study was conducted in which the brand of car was held constant and the accessibility of knowledge that could be used to think of reasons varied—specifically, some participants were shown an advertisement listing features of a Toyota before they were asked to think of either one or ten reasons to drive a BMW. This method was based on the hypothesis that exposure to the Toyota ad would make the task of generating many reasons to drive a BMW seem easy and might even help consumers accomplish the task. If so, then participants who saw the Toyota ad would view retrieval ease as uninformative and instead rely on the content they considered, leading to a more favorable evaluation of BMW when ten reasons rather than one was requested. Participants who were not shown the Toyota ad were expected to rely on retrieval ease and report a more favorable evaluation of BMW when they were asked to think of one rather than ten reasons, as they had in the initial studies. The findings were consistent with these predictions.

The findings of Tybout and her colleagues suggest that practitioners should exercise caution when developing messages that seek to engage the audience by challenging people to generate reasons to buy their products. The optimal number of reasons that should be requested may vary as a function of the target audience’s knowledge base. Requesting a large rather than a small number of reasons may increase liking for the product if the target audience is composed of either novices or experts. However, this strategy may backfire if the target audience has moderate knowledge. The authors’ research also suggests an approach that may succeed regardless of the level of knowledge. In one of their studies, they show that playfully asking participants to “imagine” reasons (rather than challenging them to think of reasons) results in equally high evaluations regardless of whether one or ten reasons are requested from moderate-knowledge participants. It seems that a more creative, genial frame makes generating even ten reasons seem easy. Thus, a subtle change in wording may make more reasons superior to fewer regardless of whether consumers think about the retrieval ease or the content. More generally, the research suggests that responses to persuasive messages are based not only on information people have about a brand but also on their reflections about the process by which their judgments were made.

Further reading:

Wänke, Michaela, Gerd Bohner, and Andreas Jurkowitsch (1997). “There Are Many Reasons to Drive a BMW: Does Imagined Ease of Argument Generation Influence Attitudes?,” Journal of Consumer Research, 24(2): 170-77.