The Value of First Impressions: Leveraging Acquisition Data for Customer Management

Firms increasingly have access to richer customer data. What a decade ago was merely a transaction added to a customer database has become a collection of behaviors that a customer engages in while she is making a purchase (e.g., whether her purchase was online or offline, whether she used a tablet or computer, whether she bought a new product or an old best-seller). We posit that these customer's decisions carry valuable information for the firm as they could potentially explain a large proportion of the heterogeneity in future behavior, both in terms of what the customer is expected to do (i.e., her lifetime value) and how responsive she will be to marketing actions. The latter point is especially relevant for contexts in which firms do not observe many purchases per individual (e.g., retail) or where targeting occurs soon after customer acquisition. In these contexts, traditional models that only rely on unobserved sources of heterogeneity are unable to help the marketer target customers with precision. 

To overcome this limitation, we propose a model that allows marketers to form customers’ “first impressions” by leveraging the data collected at the acquisition moment—data that already reside in the firm’s database and are available for every customer. The main aspect of the model is that it captures latent dimensions that impact both the variety of behaviors collected at acquisition, as well as future propensities to buy and to respond to marketing actions. Using probabilistic machine learning, we combine deep exponential families with the demand model, flexibly relating first impressions with consequent customer behavior. We first demonstrate that such a model is flexible enough to capture a wide range of heterogeneity structures (both linear and non-linear), thus being applicable to a variety of behaviors and contexts. We then applying the model to data from a retail context and illustrate how the focal firm could form customers’ first impressions. We show that the focal firm would significantly improve the return on their marketing actions if it targeted just-acquired customers based on their first impressions.