How to project customer retention

Abstract

How to Project Customer RetentionAt the heart of any contractual or subscription-oriented business model is the notion of the retention rate. An important managerial task is to take a series of past retention numbers for a given group of customers and project them into the future in order to make more accurate predictions about customer tenure, lifetime value, and so on. In this paper we reanalyze data from a leading book on data mining (Berry and Linoff 2004), who drew the dire conclusion that "parametric approaches do not work" for such a task. As an alternative to common "curve-fitting" regression models, we develop and demonstrate a probability model with a wellgrounded "story" for the churn process. We show that our basic model (known as a "shiftedbeta-geometric") can be implemented in a simple Microsoft Excel spreadsheet and provides remarkably accurate forecasts and other useful diagnostics about customer retention. We provide a detailed appendix covering the implementation details and offer additional pointers to other related models.

Improving profitability with customer‐centric strategies: the case of a mobile content provider

“…More specifi cally, the telephone operator as a source of customers can be considered a proxy of income. Moreover, a longer length of subscription is a proxy of customer satisfaction, since it is well known in the literature that the longer subscribers are with the company, the lower the probability that they are going to churn (Fader and Hardie, 2007;Fader and Hardie, 2010;Gupta and Lehmann, 2005). Regarding the relation between marketing campaign and lifetime, as we already mentioned, diff erent channels attract customers with diff erent retention behavior.…”

Customer-Base Valuation in a Contractual Setting: The Perils of Ignoring Heterogeneity

“…Given (1), it follows that the retention rate associated with the sBG model is Fader and Hardie (2007a) stop with the development and empirical testing of the sBG model. Now we take it to the next step by showing how to generate estimates of CLV given the model's parameters.…”

“…We now wish to undertake a more systematic study of this problem, still assuming that the dynamics in cohort-level retention rates stem entirely from cross-sectional heterogeneity. We will undertake this analysis using the shifted-beta-geometric (sBG) model for contract duration proposed and tested by Fader and Hardie (2007a). The sBG model is based on the following two assumptions:…”

“…In §2, we present an exploratory analysis in which this phenomenon is simply an artifact of crosssectional heterogeneity in individual retention probabilities (e.g., Gupta and Lehmann 2005, pp. 29-31); this is examined more formally and systematically in §3 using the shifted-beta-geometric model of contract duration (Fader and Hardie 2007a). We then explore the implications of failing to account for retention dynamics on the true magnitude of "retention elasticities."…”

“…As we typically observe increasing cohort-level retention rates (e.g., Fader and Hardie 2007a, Kumar and Reinartz 2006, Reichheld 1996, Schweidel et al 2008, it is crucial that our calculations incorporate this phenomenon.…”