Strong consumer behavior research, like any primary research project, begins with clear research objectives. Once those objectives are documented and agreed upon, developing hypotheses helps us make informed research design decisions. Thoughtful hypotheses can sharpen our scope and help us determine which methods (qualitative, quantitative, or a mix) will best address the study’s objectives.
So where do useful hypotheses about consumer behavior come from? Customer insights, CX, and UX researchers commonly draw on several sources, including these three:
- Secondary research. Prior in-house studies, analyst reports, syndicated research, and academic literature can help us identify and describe relevant behaviors.
- Exploratory primary research. Customer or expert in-depth interviews (IDIs), shopper observations, and analysis of social media/online product reviews can inform hypothesis development. Even a small-scale study can provide fresh learnings about the behaviors we want to understand.
- Established behavioral models. Behavioral science models are widely used by Fortune 100 companies and other organizations to identify and categorize drivers of, and barriers to, consumer behavior. These frameworks help us develop hypotheses that might challenge existing assumptions about a target population’s shopping, media consumption, product usage, or other behaviors.
Our recent Live Event, The Art & Science of Hypothesis Generation: Using Behavioral Science Models to Spark Better Consumer Insights, focused on the third source: established behavioral models. Instructor Jeff McKenna explored how two widely respected models, COM-B and the Fogg Behavior Model, can help researchers develop more informed hypotheses and translate them into research design choices. Following is a summary of Jeff’s presentation, along with related resources.
First, clarify the consumer behavior we want to understand
A behavior is an action someone takes. In the context of consumer research, that often means we are exploring “actions” (behaviors) related to shopping, product usage, technology adoption, media consumption, or even lifestyle activities.
Specificity matters. “Healthy eating” describes a broad topic, while “a shopper choosing a healthier snack during a rushed grocery trip” identifies an action within a particular context. That context gives us something concrete to investigate.
What makes a hypothesis useful?
A hypothesis is a proposed, testable explanation or prediction. It proposes something that research can support or refute. Avoid the trap of confusing a research question with a hypothesis:
- Research question: “What prevents shoppers from choosing our healthier snacks?”
- Hypothesis: “Difficulty identifying snacks that meet their health goals reduces shoppers’ likelihood of selecting those products.”
A useful research plan also considers competing explanations. Perhaps shoppers understand the products perfectly well but cannot find them, dislike their taste, or routinely reach for something familiar. Exploring multiple hypotheses helps us avoid designing an entire study around the first explanation that appears plausible.
How behavioral science broadens the hypothesis set
Generating alternative hypotheses takes deliberate effort, especially when a team already has a well-established view of its customers. Behavioral science offers frameworks for examining influences that can be easy to overlook, including environmental constraints, habits, and the circumstances surrounding a decision.
The Live Event highlighted COM-B and the Fogg Behavior Model, two widely used models. They are among several behavioral frameworks researchers can draw on to inform our thinking about drivers and deterrents of consumer behaviors of interest.
COM-B: Capability, opportunity, and motivation
COM-B examines how capability, opportunity, and motivation interact to influence behavior. Applied to the session’s hypothetical healthier-snack example, its three categories suggest different explanations for disappointing sales:
- Capability includes the physical and psychological capacity to perform an action, such as relevant knowledge and skills. We might hypothesize that shoppers struggle to interpret nutrition information and therefore have difficulty identifying products that meet their goals. A product-identification task could reveal whether that difficulty actually exists.
- Opportunity concerns the physical and social circumstances surrounding an action. We might hypothesize that healthier snacks are unavailable at the places or times when shoppers usually purchase food. An availability audit and shopping observations would help investigate this explanation.
- Motivation includes conscious goals and intentions alongside automatic processes such as habits and emotions. We might hypothesize that a familiar snack is the default choice during rushed shopping trips, even among consumers who express strong interest in healthier eating. Purchase diaries and repeated observations could help examine that pattern, although repeated purchasing alone would not establish habit as its cause.
COM-B gives us a way to investigate how these influences work together. A shopper could understand healthier options and want to buy them, yet encounter limited availability during the shopping occasions that matter.
Fogg Behavior Model: Motivation, ability, and a prompt
The Fogg Behavior Model proposes that behavior occurs when sufficient motivation and ability converge with a prompt (trigger) at the same moment. Its emphasis on the moment of action gives us another way to investigate the gap between intention and behavior. Examples:
- Motivation: How much does the shopper want the healthier option at that moment?
- Ability: How easy is the purchase, given the time, effort, and resources required?
- Prompt: What cues the shopper to act?
A prompt can be a reminder, an environmental cue, or an event within an established routine. It does not have to be a marketing message.
For example, we might hypothesize that a relevant shelf cue increases selection of a healthier snack among shoppers who already want it and can easily purchase it. An experiment comparing observed choices with and without that cue could test the prediction.
Fogg’s ability emphasizes how easy an action is to perform. COM-B distinguishes personal capability from external opportunity. These concepts overlap in practical application, but their meanings and roles within the models differ.
How hypotheses improve research design
The practical benefit becomes clear when we begin designing a study. Rather than collecting broad opinions about healthy eating and hoping an explanation emerges, we can use one of these models to identify and categorize the hypothesized drivers of, and barriers to, the behavior we’re studying. Because the model prompts us to consider a sufficiently diverse range of possible influences, it reduces our reliance on existing assumptions. It also helps us choose appropriate methods and write specific questions and tasks for data collection.
For example, asking “How important is healthy eating to you?” tells us about a stated priority. It doesn’t tell us what helps or prevents someone from actually eating healthily: whether they can interpret a label, find a suitable product, or notice it during a rushed shopping trip (all potential barriers to the desired behavior).
For market research and insights professionals, behavioral science models such as COM-B and the Fogg model, are proven, practical tools for research planning and design.
Related Resources
- Michie, van Stralen & West (2011). The Behaviour Change Wheel. Implementation Science 6:42.
- Fogg (2009). A Behavior Model for Persuasive Design. Persuasive 2009.
- Behavioural Insights Team (2014). EAST: Four Simple Ways to Apply Behavioural Insights.
- Wood & Rünger (2016). Psychology of Habit. Annual Review of Psychology.
