Customer Behavior and Class Actions

In this Q&A, John R. Hauser, the Kirin Professor of Marketing at MIT’s Sloan School of Management, and an Analysis Group affiliate, explains “customer heterogeneity,” a marketing concept that can be applied usefully in litigation matters, particularly in the area of class certification.

Q: What is “customer heterogeneity,” and why is it relevant in litigation?

Professor Hauser: Customer heterogeneity is a very intuitive concept that refers to how consumers differ from one another in their demographics, attitudes, behaviors, and, of course, preferences for products. Each of us typically thinks of a given product as a bundle of different features and services that collectively meet our needs in various ways, but we also consider different aspects of the product as more or less critical to our purchase decision.

My colleagues and I have researched ways to accurately evaluate consumer purchasing behavior for existing and as-of-yet-undeveloped products and determine the most relevant drivers of consumer decisions, even if differences among consumers exist. Companies can use this information to design, price, and market new products and make accurate predictions about who will buy what. That is the business application. However, I have provided expert evaluation of consumer preferences and heterogeneity in a wide array of legal cases involving, for instance, the determination of class certification, assessments of royalty rates and lost profits in patent-infringement cases (including evaluations of the entire market value rule) and examinations of the effects of allegedly misleading statements in false advertising cases.

Q: How is heterogeneity relevant in class certification?

Professor Hauser: Let’s consider a case where consumers allege false advertising related to product performance. An analysis of customer heterogeneity, and of several forms of surveys, can be useful in accurately assessing numerosity, commonality, and conflict. For example, a survey may reveal that only a handful of consumers care about the feature in question because the circumstances in which it is relevant are rare. In such cases, a defendant might use heterogeneity and the infrequent consideration of the feature in question to defeat class certification.

Similarly, an analysis might suggest that for most or even all consumers a feature was a primary driver in their decision making and is of great value to them. In that case, a plaintiff might use the lack of heterogeneity to support class certification. Finally, an evaluation of consumer heterogeneity might reveal consistent or inconsistent preferences for particular remedies. If such conflicts among plaintiffs exist, the defendant may be able to defeat the class before it is certified.

Q: Can you offer an example?

Professor Hauser: I remember a case in which a class was dismissed due to customer heterogeneity, specifically as it related to preferred remedies. A class action was brought against manufacturers of pressure-treated woods, which were used by a range of private and professional customers in different kinds of structures. I conducted a survey that examined typical usage scenarios, the drivers of the purchasing decision, and the potential alternatives available to members of the proposed class. The results demonstrated extreme heterogeneity in impact and usage. Furthermore, preferred remedies varied enormously among potential class members, with a few seeking remedies proposed by the plaintiffs and others not; others were indifferent to the performance issues. Consequently, the class wasn’t certified.

Meanwhile, in a consumer goods case, I observed that consumers were consistent in saying they valued an attribute like “healthy,” but inconsistent on how much they valued that attribute. In that case, it was important to distinguish between heterogeneity sufficient to defeat a class and expected heterogeneity. The latter is about variation in the marginal value of a product or attribute along a demand curve, which can be observed in nearly every market. In such cases, it is possible to evaluate expected variation in the specific value a consumer places on an attribute, while still demonstrating that most purchasers consider it an important attribute. ■