After more than three decades in marketing research, on both the supplier and client side, I have concluded that my profession is greatly misunderstood. Its ability to discover truth is imperfect, and the task inherent in predicating consumer behavior is formidable, yet it is often thought of as a “simple” solution.
How did this happen?
Many users think that marketing research can easily answer their questions with a few focus groups followed by a quick Internet survey. “Just ask consumers X about subject Y, so we can make decision Z.”
VOC—Voice of the Customer—becomes management’s guiding star, an acronym that belies the sophisticated nature of marketing research as a social science. This view, if I may be allowed a brief emotional outburst in what I hope is a more dispassionate essay, drives me nuts!
Chief among the reasons for all the misunderstanding are the following:
Lack of objective reality: Something has objective reality if people can agree to its existence. Researchers try to measure stuff going on in consumers’ minds that we feel influences their market behavior. But what is this “stuff?” Short of getting into a philosophical discussion, we really don’t know. So researchers develop constructs like “satisfaction,” “intent to purchase,” “quality,” etc. We believe if we ask consumers questions about these constructs—“How satisfied are you with boat brand X on a scale from 0 to 10—we can use their answers to predict behavior. Sometimes this works; sometimes not. For example, consumers don’t always re-purchase a product even if they are extremely satisfied. Their needs and wants may change, and different products may suit them better.
Consumers don’t know: Several empirical studies support the proposition that, for some decisions, consumers simply have difficulty articulating why they purchased what they did. This can occur if the decision process is highly complicated and involves rational as well as emotional reasons underlying a purchase. Additionally, consumers often have difficulty identifying the difference between a product attribute that is merely desirable (e.g., a driver-side airbag, which all cars have, so, therefore, cannot be a determinant of behavior) vs. a product attribute that is both desirable and a determinant of behavior (e.g., the styling of a car, which differs in varying degrees).
Respondent contamination: If we ask a consumer to rate a product on a specific attribute, or to discuss the role a particular issue played in a recent purchase decision, we can unintentionally lead a respondent to focus on an issue that played no role in the decision at all.
We can’t measure everything: Marketing researchers make an effort to develop statistical models that predict either actual purchasing behavior or purchasing intentions. But these models cannot account for all the factors that might affect behavior such as a competitor’s market response, the quality of the marketing communications supporting a product launch, or the final quality of the product delivered to the market.
Consumers may lie: Why? One reason is that they often tend give a socially acceptable answer. This is especially true in food research where consumers might say they want to purchase healthy products when, in reality, they don’t. Some consumers don’t want to admit they purchased a particular brand because they want to project a certain image to their friends.
All of these factors confound our ability to discover the truth about the consumer. But don’t lose hope. Marketing research does work, although imperfectly. And improved measurement and tools have overcome many common pitfalls.
One field of study that aids our ability to measure and understand consumer behavior is psychometrics, which examines issues related to measurement validity of the questions we ask respondents. Some ways of asking questions are more valid than others, and more valid questionnaire designs provide more valid measures of constructs such as perceived product quality and brand loyalty – critical to understanding consumer behavior.
On another front, academics and applied researchers have made major strides in constructing statistical models that researchers use to gain more valid insights into factors affecting the consumer decision process. One example is “marketing mix” models, which attempt to understand how various marketing expenditures affect consumer behavior.
These new and better tools work. Current literature is filled with hundreds of peer-reviewed articles showing how companies using these new methods and techniques, vs. those that don’t, are more successful and profitable. Admittedly, organizations attempting to optimize the validity of their marketing research typically employ senior managers who embrace the marketing concept, which accounts in part for these successes.
Bottom line: First, don’t short change quality marketing research. A penny saved may be a marketing misfire costing your organization millions of dollars in direct costs and untold benefits in foregone opportunities. Second, remember the wise words of advertising guru David Ogilvy: “I notice increasing reluctance on the part of marketing executives to use judgment; they are coming to rely too much on research, and they use it as a drunkard uses a lamp post—for support, rather than for illumination.”