“Counting Your Customers” the Easy Way: An Alternative to the Pareto/NBD Model

Abstract

"Counting Your Customers" the Easy Way: An Alternative to the Pareto/NBD ModelToday's managers are very interested in predicting the future purchasing patterns of their customers, which can then serve as an input into "lifetime value" calculations. Among the models that provide such capabilities, the Pareto/NBD "Counting Your Customers" framework proposed by Schmittlein, Morrison, and Colombo (1987) is highly regarded. But despite the respect it has earned, it has proven to be a difficult model to implement, particularly because of computational challenges associated with parameter estimation.We develop a new model, the beta-geometric/NBD (BG/NBD), which represents a slight variation in the behavioral "story" associated with the Pareto/NBD, but it is vastly easier to implement. We show, for instance, how its parameters can be obtained quite easily in Microsoft Excel. The two models yield very similar results in a wide variety of purchasing environments, leading us to suggest that the BG/NBD could be viewed as an attractive alternative to the Pareto/NBD in most applications.

Customer Lifetime Value Prediction in Non-Contractual Freemium Settings: Chasing High-Value Users Using Deep Neural Networks and SMOTE

“…The marketing literature abounds of conceptual [6,7], methodological [13,14] and empirical [15,16] studies of customers' LTV. Methodologically, most empirical contributions are routed in stochastic models of customer behavior [10,14,16] or regression approaches [15,17].…”

“…Methodologically, most empirical contributions are routed in stochastic models of customer behavior [10,14,16] or regression approaches [15,17]. The datasets at use mostly describe monetary transactions of customers.…”

“…Data are hence highly imbalanced [25] and there is abundant non-purchase related data that describe user behavior in depth and that can be valuable for prediction [1]. The stochastic models proposed by the marketing literature [14] are somewhat ill-equipped to deal with this kind of data as they were conceived for purchase related data. It would certainly be interesting to extend them to highly detailed usage data, but this is beyond the scope of the current study -especially as their performance in applied settings remains uncertain [26].…”

“Counting Your Customers” One by One: A Hierarchical Bayes Extension to the Pareto/NBD Model

“…Further, λ and μ follow independent gamma distributions, which are formulated as a mixture distribution model. Although their work is highly regarded and follow-up research has been conducted (Fader, Hardie, andLee 2005a, 2005b;Reinartz andKumar 2000, 2003;Schmittlen and Peterson 1994), it is the increasing importance of new types of marketing, such as Database Marketing, CRM, and One-to-One Marketing, that has brought this model to the attention of researchers and practitioners.…”

A methodology based on profitability criteria for defining the partial defection of customers in non-contractual settings

“…The negative binomial distribution (NBD) model by Ehrenberg (1959) was the starting point for modelling the repeated purchase behaviour. One of the first references proposing a NBD model is the work by Schmittlein et al (1987); extensions and empirical validations of this model can be found in Schmittlein & Peterson (1994) or Fader et al (2005), while applications can be found for example in Reinartz & Kumar (2000, 2003 and Hopmann & Thede (2005). Also with a stochastic approach, Netzer et al (2008) apply a Hidden Markov model in order to capture the dynamics of the behaviour of the customers (donors of several universities).…”