Careful analysis of marketing decisions can play a significant role in litigation, particularly in class action matters at the class certification stage. Randolph Bucklin, Professor of Marketing, Peter W. Mullin Chair in Management at the UCLA Anderson School of Management, has served as an expert witness in numerous cases over the past 15 years, including several class action litigations. Analysis Group Vice President Peter Simon spoke with Professor Bucklin about the role that quantitative analysis has played in evaluating purchasing decisions and other behavior by consumers (and businesses) in several notable recent cases.
Dr. Simon: How is your research background helpful in a litigation context?
Analysis Group Affiliate Randolph Bucklin
Professor Bucklin: My research experience has shown me that, although there are generally known motivations for consumers’ and firms’ behaviors, there is in fact a very wide variety of reasons for the specific decisions made by individual consumers and businesses. That perspective often brings to light factors that others have not considered, or have glossed over, when evaluating interactions between consumers and businesses. Accounting for these missed factors has the potential to overturn conclusions based on a more simplified view.
This background is particularly helpful when analyzing issues in class action litigation at the class certification stage. In these cases, it’s natural to think about how consumer or firm behavior drives decisions. Sometimes, it’s a straightforward matter of understanding how the alleged “bad act” affects consumers or firms. In many other cases, however, it’s useful to explore more deeply and understand differences in conduct across different groups, or types, of consumers or firms. Doing this sometimes requires obtaining information sufficient to identify those different groups of consumers or firms – typically data on market participants’ individual circumstances – that can be used to discern group differences.
My academic research over the years and my experience working with data has helped me to take advantage of the growing information available about consumers and firms. In particular, I have conducted research using, and have extensive experience with, data on individual consumer purchases, including very detailed material such as UPC scanner panels or Internet clickstream data. Analysis of such relevant, detailed data illustrates that marketplace decisions are not always as simple as sometimes claimed.
Dr. Simon: What are some recent examples of cases where this type of analysis was particularly impactful in the class certification phase?
Analysis Group Vice President Peter Simon
Professor Bucklin: In the Fraley v. Facebook case, I used this type of analysis to critique an opposing expert’s report. That case involved claims regarding Facebook’s use of members' personal content and activity information (such as "liking" a product or service) without members' consent to create "Sponsored Stories" content about their activity. As part of my analysis, I examined detailed data on the Sponsored Stories in which the class’s named plaintiffs appeared. With these data, I was able to show that, contrary to plaintiffs’ theory, other displayed content sometimes performed better than Sponsored Stories for some plaintiffs. This highlighted the need for individual inquiry to draw conclusions regarding class members. That’s an example of the way data analysis can be used to support an argument in class certification.
In another case, some of my opinions drew from my understanding of the details of consumer decision-making. During a recent project with Analysis Group – Dang v. San Francisco Forty Niners – an opposing expert had estimated, through the use of a survey, how consumers’ purchase decisions might have changed in response to a hypothetical price change. I pointed out that the opposing expert’s estimates were flawed because they were based on only one aspect of consumers’ purchase decisions, and that many other well-established factors might have affected decisions in potentially countervailing ways. For example, the opposing expert did not consider the effect from advertising or the ways in which products were distributed. Thus, the extent to which those factors might have influenced consumers’ reactions to a simultaneous hypothetical price change was not considered.
Dr. Simon: How does your perspective come to bear in class actions involving business plaintiffs?
Professor Bucklin: The Google AdWords case is a good example. That case involved a class composed of businesses. In that case, I explained the factors affecting the value received by advertisers on a website like Google’s. One interesting observation from that case was that it matters whether the class is composed of consumers or firms. The decision-making you’re dealing with is very different if you’re analyzing 20,000 consumers versus a group of firms, each of which may have fundamentally different factors affecting the performance of its online advertising. As a simple example, one firm may only advertise online while another advertises mostly through broadcast media (television, radio) and has a very small Internet advertising presence. As a consequence, the value derived from online advertising may be very different for those two firms.
I was also able to analyze data on ads placed on at-issue web pages, and used my results to criticize an opposing expert. The expert proposed a method for calculating restitution to firms advertising through Google that relied on an average webpage performance measure. I showed that this method would have awarded restitution to advertisers whose ads’ performance exceeded the average of ads on google.com. The opposing expert assumed advertisers could all be treated equally when calculating restitution, but I was able to show that some advertisers benefited from having their ads on at-issue web pages. ■