Editor’s Intro: I’ve always felt puzzled by the attitude of many market researchers have towards “science”. After all, what we do is “research”, so one would think that market researchers would wholly embrace the scientific method and all that it entails. Yet many are openly indifferent towards or ignorant of what I would call “scientific thinking”, which connotes much more than the often esoteric “marketing science” of statisticians and modelers. Garrett’s article makes a strong argument for emphasizing scientific thinking in market research. I think it is more important now than ever.
There is a lot of discussion in the market research community surrounding the application of behavioral science to survey research. This dialogue has challenged the insights community to address several methodological shortcomings that have long plagued the field. Such weaknesses include a general lack of predictive validity, and the inability to isolate and replicate insights outside of the survey context.
To researchers, these pitfalls often reflect the status quo. To a client, these caveats determine whether a decision based on consumer insights will become a success, or fail with spectacular fashion. The interjection of behavioral science has challenged both parties to think critically about the insight collection process and the application of valuable learnings to solve business challenges.
This change in vantage point is spurring innovation in areas that have traditionally gone unchallenged since their inception. The namesake field of behavioral science has lent the most impact to market research, more so in the rigorous approach than any specific methodologies used. At its core, behavioral science utilizes the scientific method to understand the drivers of behavior and introduce theories to guide an observed behavior towards a targeted alternative.
On a broad scale, behavioral insights are applied to reduce the consumption of unhealthy food, increase the amount of those who sign up for organ donation, and ensure that retirement funds are endowed well for the future. In the consumer context, behavioral science is used to increase the adoption or uptake of new products, and strengthen the relationship that consumers have with brands.
As the discourse on applying behavioral science to enhance the understanding of consumer behavior continues, there are three things that market researchers can do to implement a more scientific approach to their work:
1. Identify a testable hypothesis
Designing quantitative questionnaires with a pre-defined, testable hypothesis in mind ensures that the insights uncovered in the study are in line with the stated business objectives. Just as scientists develop hypotheses before testing theories in a lab setting, market researchers should work cooperatively with their clients to identify the underlying factors that answer the fundamental business question. Defining the business problem upfront and ideating solutions supports the creation of clear and concise deliverables that are actionable. Reframing research briefs from “what would we like to uncover” to “what sort of behavior do we think we can change” brings traditional market research one step closer to its social science brethren.
2. Measure actions, not an intent
Consumers struggle with predicting their behavior thirty, sixty, or ninety days into the future. This fact challenges the convention of traditional survey logic where researchers use their one touch point with consumers as a be-all, end-all data capturing moment. One way to circumvent this future prediction inability is to ask survey respondents questions that they can answer at the moment. Asking consumers to choose one product over the other in a forced choice fashion, or to “invest” game-like currency into one concept or another starts to replicate how consumer choices are made in the real world, rather than evaluating them in a speculative bubble.
3. Utilize control groups
Common reactions to this recommendation are often immediately followed by questions of: Use a control group? For what? How will that enhance my current concept testing study design? When testing concepts of any sort, utilizing control groups isolates the potential effects that a new concept may have on the parent brands’ perception. Without an unexposed control group, it is nearly impossible to identify the extent to which the concept alters brand perceptions. Control groups are a valuable addition that can help explain how well one concept iteration performs over another and the impact that it has on the brand.
There are countless crossover applications to market research one can borrow from behavioral science. The underlying scientific method-based approach ensures that there are accountability and confirmatory powers in consumer research; two components that researchers have struggled achieving in recent years when utilizing the present approach.
The first step in uncovering actionable insights is ensuring that the data patterns discovered are representative of human behavior in the marketplace. Additionally, insights should be easily replicated across large populations to ensure that the responses observed reflect how a majority of consumers will act in a given situation. Methods used in traditional online surveys should not be asked in an isolated research bubble either; instead, successful researchers must employ tools that bring the insights uncovered closer to the reality of how consumers make decisions.
Applying the three takeaways listed above is not the final chapter to this story; instead, it is where the challenge to expand the horizon of market research begins.