20 Oct 2021

Behind the Research: Eva Ascarza


by Shona Simkin

Eva Ascarza

Eva Ascarza is the Jakurski Family Associate Professor of Business Administration in the Marketing Unit, and teaches Marketing in the MBA required curriculum. We asked Eva about her research and what makes her happy when she is not teaching and working on data modeling.

How did you become interested in your area of study?
I am a mathematician by training—I like numbers, data, and to develop methodology that can make sense of data to answer relevant questions. I ended up in marketing accidentally. I got an internship at a bank after college, and was placed as a junior person in a newly-formed customer relationship management (CRM) group. But no one else was in the group—no senior person, just me—and I had no background in marketing or anything, so I ended up spending my time there reading books and papers (about CRM), and helping my senior co-workers (in other departments) with numbers in excel. Granted, I didn’t do much CRM there, but I realized that I liked research and had a curious mind, so I applied to PhDs in marketing.

The first few classes I took in the PhD program were about probability models, and how the patterns of behavior that we see in the market are merely an aggregation of many different individual customers making decisions—customers with some common rules of behavior, but with different preferences, different circumstances... That truly fascinated me. I could write a few formulae based on how I believed customers behave; essentially, translating “common sense” to math, then fitting the data to my equations, and voila!, now I can explain patterns that were hidden in the data, and I can also predict what these customers would do next. I realized there were a lot of things to do in helping firms figure out how to better understand and manage a customer.

What is the focus of your research?
I still look at how to better leverage data to understand behavioral patterns and get insights to help companies manage their customers. Compared to when I started, now I have a much better sense of which problems I want to help fix and research. I combine collaborations with firms with problems that I think the academic community can benefit from solving—issues for which there is no current approach, insight, or model.

Some of my ideas or findings are counterintuitive at first, but convincing once you explain it; it is common sense applied to reality. Plus, when you show it empirically, and in multiple contexts, firms listen and (hopefully) change their behavior.

For example, I was working with different companies on how they retain customers, and the most common approach was putting lots of data to the machine learning models and figuring out who is the customer most likely to leave—this is called “churn.” And I thought, “Wait, targeting only the people at the very top of the list seems useless, because many of these people are going to leave anyway, and you might also be missing several others who would stay if you try keeping them.” All they're doing is figuring out who is going to leave, not who they can keep. It's obvious that the latter would make more sense if you want to increase retention. Yet, everyone was predicting churn. I didn't find any papers that were making that point empirically, so I reached out to firms that I knew were running retention experiments, and two companies gave me their data. I analyzed their data and showed that they would be better off working to identify the customers who could be persuaded to stay, regardless of their probability to churn.

What are you working on now?
I have several projects at the moment, mostly on the topic of personalization. I'm most excited about work that I’ve been doing with my colleague Ayelet Israeli. It all started when we prepared some materials to teach students about algorithmic bias. We created typical data from a marketing campaign and gave it to the students to analyze. Their task was to identify which customers should be targeted in the next campaign for the company to maximize profits, which students can do relatively easily. Then, we reveal the demographic information of those customers, only to realize that targeted customers are more likely to be white (than non-white) and male (than non-male). Now what? Companies want to increase profit, but what if your objective is also to be fair to the people you target?

Solving this situation is not easy, and that’s exactly what we do in our research. We have developed a solution (in more nerdy terms, an algorithm) that allows firms to balance both objectives, profitability and fairness, when they personalize their policies. We are now in the process of applying our method to different empirical contexts, broadening their impact beyond marketing.

What do you like to do in your spare time—what makes you happy outside of your research?
Jamón—Spanish ham—makes me very happy. And my Vespa! I take rides on my Vespa and it gives me a smile on my face. And definitely seeing my childhood friends and family. I am from Pamplona, a small city in the north of Spain. Before COVID I would go to Spain even for a weekend—when I am there I go back to my origins, not only geographical but behavioral and mentally speaking; it’s like renewing my soul (laughs). So, I guess jamón, red wine, Vespa, and friends are my answer. Not necessarily in that order! In my spare time I also exercise, or try to, but only because I think of my future self and that she would appreciate it—it's an investment in “future Eva" and also the excuse to sometimes switch a red wine for a Negroni. That's how I constantly deal with myself.

Read more about Eva Ascarza in Working Knowledge. For updates on HBS faculty research, sign up for Working Knowledge’s weekly e-mail newsletter.

Post a Comment

Comments must be on-topic and civil in tone (with no name calling or personal attacks). Any promotional language or urls will be removed immediately. Your comment may be edited for clarity and length.