Written by Lina Altenburg, Ph.D. candidate at KU Leuven (Belgium)
Who would have thought that an interview about a paper that develops a method to determine whether customers become inactive in non-contractual settings would end with an AI-generated song text about control charts? But let me start from the beginning.
I had the chance to interview Niels Holtrop (Assistant Professor at Maastricht University, the Netherlands) and Jaap Wieringa (Full Professor at the University of Groningen, the Netherlands) about their paper “Timing customer reactivation initiatives”, which was published in the September 2023 issue of the International Journal of Research in Marketing.
In the article, Niels and Jaap identify which customers are inactive and when a customer becomes inactive. They use this knowledge to develop an approach that helps companies send customer reactivation emails at the right time and to the right customers. The authors collaborated with a greeting card and gifts company that was unsatisfied with its customer reactivation strategy. Niels and Jaap used the historical data of the company to develop a suitable methodology and then implemented their adjusted reactivation strategy in a field experiment to test the extent to which it improved the company’s previous customer reactivation performance.
Don't wake the sleeping dogs
The project started as part of Niels’ PhD dissertation at the University of Groningen. This was his third Ph.D. research project together with Jaap; a “forced collaboration” and they “were doomed to this project” as they state cheerfully. The two have known each other since Niels’ student times at the University of Groningen. A few years later, Jaap became Niels’ PhD advisor.
Niels was interested in a research topic related to customer management and Jaap had a participant in his executive training course that was particularly interested in customer reactivation and whose company was willing to provide data to investigate this topic. As Niels puts it, “the stars aligned” for this project.
“How can we reactivate these customers [...] while, well, making sure that we don't wake the sleeping dogs, I think that was the initial idea that we had.”
Niels Holtrop
The main issue that the company was struggling with was that they did not know when exactly to reactivate customers, and which ones to reactivate. They had an internal benchmark of 8 weeks after which they would send a reactivation email, but they also knew that the length of this period was not suitable for customers that had much shorter or longer inter-purchase times. Based on the literature, Niels and Jaap knew that it can also hurt to send reactivation emails to the wrong customers at the wrong time (i.e., wake the sleeping dogs), as they might not react favorably to the company’s communication. Hence, the research project was born.
An old love
The authors use a methodology based on control charts to analyze which customers become inactive at what moment. Control charts are a methodology that has been used in statistical quality control. They are suitable for determining customers’ inactivity as they allow for identifying deviations from normal variations in the customers’ inter-purchase time. In case a control chart indicates that the purchase time becomes unusually long (i.e., exceeds the limits of normal variation), this can be seen as a possible signal that the customer became inactive. Niels and Jaap also integrated customer heterogeneity in the control charts, allowing for individual differences between customers, as customers can differ largely in their average inter-purchase time (i.e., some customers typically buy every other week, others may buy only once or twice per year).
A simple example of a control chart provided in the paper by Niels and Jaap
The authors’ choice for using the control chart as the underlying methodology of their analysis was not random. Jaap's PhD thesis in the field of statistical quality control (at the University of Groningen) was on control charts. For some time, he was already trying to marry his old love (control charts) with his new love (marketing), as he puts it. And this project was the perfect opportunity to do so.
“This was a project that was long on my wish list to do”
Jaap Wieringa
Hence, the choice of the methodology was already quite clear from the beginning. This also didn’t change during the review process. The reviewers did challenge Jaap and Niels to not just think in terms of signal (i.e., which customer should be targeted when) but also to relate their findings to performance measures (i.e., how much better is it to other approaches and how much additional profit can it generate). Hence, they put a marketing lens on the engineering tool.
Can you sing about control charts?
Okay, but how does this whole discussion about control charts and customer reactivation lead to AI-generating lyrics for a song? At the end of our interviews for IJRM, we often ask the authors some personal questions. When I learned that Jaap had been active as a singer in a choir and his career could have turned into a music career, I told him that he made the right career choice because “I don’t think you can sing about control charts” (i.e., his old love). Well, the two proved me wrong. With the help of AI, Niels generated the lyrics for a song about control charts. Read the whole song below. I guess we have a new way of promoting the papers published at IJRM...
Read the paper
Interested in reading all the details about how control charts can help with the timing of customer reactivation emails? Read the full paper here.
Want to cite the paper?
Holtrop, N., & Wieringa, J. E. (2023). Timing customer reactivation initiatives. International Journal of Research in Marketing, 40(3), 570-589.
Meet The Authors
Niels Holtrop, Assistant Professor of Marketing at Maastricht University (the Netherlands)
What drives you to do the research / work that you do?
I think Marketing is very tangible. Marketing has a strong presence in our daily lives and I think that's nice because it gives you inspiration for research. You can also apply what you read about in real life, or change your own behavior based on what you read.
If you were not in academia, what would you be?
I would probably have some kind of data scientist role. Because that's what I really like. I really like analyzing data and coming up with insights from that. I think that is what I do academically, so probably I would also do that professionally, but I don't know. I don't have the counterfactual.
Jaap Wieringa Full professor of Research Methods in Business at the University of Groningen (the Netherlands)
What drives you to do the research/work that you do?
For me, that is curiosity. If I see something that does not work in practice, that makes me wonder why and go from there. That is what is driving me. I also like to work on things that are relevant to the field. I think marketing is a very nice field because it relates to so many theoretical fields like economics or psychology, sociology, and business administration. So, from the theoretical perspective many fields are relevant to the marketing area. The same holds when you take a methodological point of view: I very much enjoy working with a broad range of techniques from econometrics and computer science and applying them to marketing problems. They all come together in this field and are relevant to the business context in which the marketing function is embedded.
If you were not in academia, what would you be?
How far do you want to go back? Because when you are in primary school, you have ideas of what you want to become later, but they typically develop over time. After finishing my PhD my initial aim was to become a consultant applying technical methodologies in practice. Because I like the interaction between the research methods and solving practical problems. Working at that interface is something that suits me.
But going back a little bit further, I'm also very much interested in the technical part of stuff. If my life would have evolved differently, I might have been working as a technician in a recording studio. Another thing is I like singing and acting, that is something I still actively do, in in a different life I might have pursued a career as a musical star. Those are all alternative career paths.
This article was written by
Lina Altenburg
Ph.D. candidate at the KU Leuven (Belgium)
Comments