Understanding Consumer Behavior in the Digital Age: A Q&A with Alix Barasch

Fast-moving technology is reshaping how consumers shop, share information, and perceive price fairness in the marketplace.

Companies are facing growing scrutiny and litigation over practices involving algorithmic pricing, customer service chatbots, and brand messages on social media. As disputes related to pricing, advertising, and consumer deception continue to emerge, it is important to understand how these technologies shape consumer perceptions and behavior.

To explore how marketing methods can be used to evaluate the consumer impact of technology developments in the marketplace, Principal Laura O’Laughlin and Manager Marimer Guevara sat down with marketing professor Alix Barasch, the Gordon and Susan Trafton Faculty Scholar at the Leeds School of Business at the University of Colorado Boulder. Professor Barasch’s research on how new technologies are reshaping consumer behavior and well-being has been published in leading journals, including the Journal of Consumer Research, the Journal of Marketing Research, and the Journal of Marketing, and has been featured in global outlets including The New York Times, The Atlantic, and NPR. Their conversation centered on Professor Barasch’s use of surveys and experimental methods to understand the consumer impact of algorithmic pricing and other new technology, as well as the implications for litigation in this area.

Your recent research examines how technology shapes consumers’ reaction to pricing information. How do consumers respond to algorithmic pricing?

In general, consumers are uncomfortable with the idea of prices being determined by an algorithm. Some consumers perceive that firms are becoming more effective at extracting maximum utility or value from their customers, which can create a sense of being manipulated or treated unfairly. At the same time, some consumers recognize that prices can vary depending on when and where a purchase is made. My favorite example of this is movie theaters charging a lot more for the same candy that is sold at the drugstore down the street. We also see this in air travel with consumers paying more for the same airline seat when purchasing closer to the date of travel.

Does consumer awareness that a retailer is utilizing algorithmic pricing negatively impact their brand perceptions or shopping behavior?

The surprising answer is not necessarily. Our research suggests that consumer perceptions of a retailer’s use of algorithmic pricing depends on how consumers interpret the fairness of the underlying price discrimination. When price differences are based on demographic characteristics of the customer – such as age, gender, or socioeconomic status – they actually prefer prices to be determined by an algorithm rather than by a human decision maker. The mechanism appears to be psychological: When demographic price discrimination comes from a human, it can feel highly personal and intentional, as though someone is specifically singling out or judging the consumer. In contrast, consumers tend to perceive algorithms as applying impersonal rules at scale rather than deliberately trying to exploit any particular individual.

Importantly, this “algorithm buffer” appears to apply specifically to demographic-based price discrimination. In contrast, when we look at pricing that varies based on more neutral factors – such as time of day or consumer demand – consumers are more uncomfortable when an algorithm determines the price. This is likely because they associate algorithmic systems with the ability to monitor consumer behavior, update prices continuously, and more effectively take advantage of consumers.

What are the implications of these findings for companies and for legal disputes involving pricing?

It is important for companies to understand that information about pricing decisions – such as whether an algorithm is involved – can meaningfully affect consumer reactions and purchase behavior. As algorithmic pricing becomes more widespread, transparency about the decision process is not just an ethical question but a strategic one.

For legal fact finders, the research has several implications. First, these findings demonstrate that consumer responses to pricing are context dependent: The same price differential may be perceived very differently depending on who or what sets it, what the price is based on, and what information is disclosed. This matters because it impacts whether consumers perceive that they were treated unfairly or disadvantaged, which can be a central question for broader claims of deception or harm.

Second, the research illustrates that claims about consumer harm from pricing practices are empirical questions, not assumptions. Whether a particular group of consumers actually felt exploited or changed their purchasing behavior depends on facts that can be measured through, for example, surveys, experiments, or behavioral studies. My research involved the careful design of experimental studies in which participants were randomized to conditions that varied in terms of the pricing agent and the context. This kind of controlled empirical design is often what is needed to establish materiality in a litigation context.

 

Across your research areas, you rely heavily on experimental and survey-based methods. How should non-experts understand the rigor and reliability of those approaches?

Experimental methods are the gold standard in psychology and consumer behavior research because they allow for causal inference. We can only have confidence that a variable caused observed changes in our outcome measure when we randomly assign participants to different experimental conditions that are exactly the same except for the variable of interest. For example, when our team wanted to know how the pricing agent influenced consumer perceptions of fairness and purchasing behavior, we used an experimental design that only varied whether the price was set by an algorithm or a human. In that way, we isolated the effect of that variable on consumer perceptions and behaviors. Because of this methodological design, our findings provide evidence of causality rather than mere correlation. In other words, they are strong evidence that the pricing agent actually drove changes in our perception and behavior measures.

When designing a survey for litigation, do you try to account for specifics of the case or claims at issue?

Absolutely. Survey-based experiments can be conducted at scale, mirroring real-world conditions and using representative samples, allowing researchers to measure outcomes that are otherwise hard to observe. Survey methods are particularly useful because they can be applied specifically to the facts of a case. A well-designed consumer survey or experiment can test whether a particular allegedly deceptive statement or omission was material to consumer decisions – that is, whether the statement or omission actually changed what consumers believed or how they behaved. Survey methods provide a more precise and defensible approach than relying on, say, assumptions or anecdotes.

Of course, the quality of any survey or experiment depends on its design. Attention to the respondent population, question framing, control conditions, and appropriate statistical analysis are all essential. As with any empirical evidence conducted in a litigation context, the methodology will be subject to scrutiny, and experts are asked to provide ample information to allow opposing parties and fact finders to properly evaluate the evidence.

 

As technology continues to evolve, what do you see as the most important open questions at the intersection of consumer behavior and the kinds of disputes that arise in litigation?

One is the expanding role of AI and automation in consumer-facing decisions. Not just pricing, but product recommendations, customer service interactions, and content curation. I’ve started to examine some of these issues in my research, for example, looking at how consumers respond differently to AI customer service agents compared to human agents across a range of contexts. Our findings suggest that a chatbot’s perceived humanness can impact consumers’ enjoyment of the interaction and their reported willingness to chat with the bot again, as well as important downstream behavior, like consumers’ willingness to share information with the company and their likelihood of taking robotic sales assistance recommendations.

I’ve also become interested in questions about transparency and disclosure. As digital environments become more sophisticated, the line between information and persuasion becomes harder to draw. I think marketing research that helps clarify when and how disclosures are effective, and when they fail to shift consumer understanding, will be increasingly relevant to both regulators and litigants. ■