Customer-base analysis seeks to use data from a firm’s customer transaction database to answer forward-looking questions such as: (1) Which individuals are most likely to be active (or inactive) customers in a future period of time? (2) What will the aggregate and disaggregate purchase patterns look like for a group of customers in the future period? (3) What is the “customer lifetime value” (CLV) of those individuals listed in the firm’s database?

While these questions are of great interest to both researchers and analysts, the standard set of statistical tools we are exposed to in our education are not well-suited for providing answers.

This two-day workshop aims to fill in these gaps by bringing researchers and practitioners fully up to speed on the essential techniques that should underlie such customer-base analysis activities. Our two main objectives are: (1) to provide an overview of the basic “building blocks” of probability models that can be used to characterise customer behaviour, and (2) to use these “building blocks” in a series of focused models/examples that illustrate several state-of-the-art approaches towards customer-base analysis (often requiring nothing more than an Excel spreadsheet for complete model implementation). By the end of the workshop, participants will understand the critical concepts and methods required to develop models for customer-base analysis and to make meaningful and accurate statements about CLV in various managerial settings. Likewise, they will be keenly aware of the limitations and concerns of other approaches that are often used for similar purposes.

This program will benefit both PhD students and marketing academics and as well as mid-level analytics professionals who want to better understand these emerging concepts. The material is somewhat technical, so some basic aptitude with probability/statistics would be beneficial for participants. For instance, it helps (but is by no means required) to have some familiarity with basic probability distributions (such as the Poisson and the binomial), even if the details are largely forgotten. Similarly, participants should be comfortable with Microsoft Excel, they do not need for any advanced capabilities (we will rely exclusively on ordinary “built-in” Excel functions). Finally, participants do not need to bring a laptop — in fact, we actively discourage it. Just bring an open mind, a sharp pencil, and a high level of interest in the subject matter. All materials presented (including the detailed spreadsheets) will be made available to all participants immediately after the seminar.