Companies in a variety of sectors are increasingly managing customer churn proactively, generally by detecting customers at the highest risk of churning and targeting retention efforts towards them. While there is a vast literature on developing churn prediction models that identify customers at the highest risk of churning, no research has investigated whether it is indeed optimal to target those individuals. Combining two field experiments with machine learning techniques, the author demonstrates that customers identified as having the highest risk of churning are not necessarily the best targets for proactive churn programs. This finding is not only contrary to common wisdom but also suggests that retention programs are sometimes futile not because firms offer the wrong incentives but because they do not apply the right targeting rules. Accordingly, firms should focus their modeling efforts on identifying the observed heterogeneity in response to the intervention and to target customers on the basis of their sensitivity to the intervention, regardless of their risk of churning. This approach is empirically demonstrated to be significantly more effective than the standard practice of targeting customers with the highest risk of churning. More broadly, the author encourages firms and researchers using randomized trials (or A/B tests) to look beyond the average effect of interventions and leverage the observed heterogeneity in customers’ response to select customer targets.

We investigate the increasingly common business setting in which companies face the possibility of both observed and unobserved customer attrition (i.e., “overt” and “silent” churn) in the same pool of customers. This is the case for many online-based services where customers have the choice to stop interacting with the firm either by formally terminating the relationship (e.g., canceling their account) or by simply ignoring all communications coming from the firm. The standard contractual versus noncontractual categorization of customer–firm relationships does not apply in such hybrid settings, which means the standard models for analyzing customer attrition do not apply. We propose a hidden Markov model (HMM)-based framework to capture silent and overt churn. We apply our modeling framework to two different contexts—a daily deal website and a performing arts organization. In contrast to previous studies that have not separated the two types of churn, we find that overt churners in these hybrid settings tend to interact more, rather than less, with the firm prior to churning; that is, in settings where both types of churn are present, a high level of activity—such as customers actively opening emails received from the firm—is not necessarily a good indicator of future engagement; rather it is associated with higher risk of overt churn. We also identify a large number of “silent churners” in both empirical applications—customers who disengage with the company very early on, rarely exhibit any type of activity, and almost never churn overtly. Furthermore, we show how the two types of churners respond very differently to the firm’s communications, implying that a common retention strategy for proactive churn management is not appropriate in these hybrid settings.