The early detection of potential churners enables companies to target these customers using specific retention actions, and subsequently increase profits. This analytical CRM (Customer Relationship Management) approach is illustrated using real-life data of a European pay-TV company. Their very high churn rate has had a devastating effect on their customer base. This paper first develops different churn-prediction models: the introduction of Markov chains in churn prediction, and a random forest model are benchmarked to a basic logistic model.
The most appropriate model is subsequently used to target those customers with a high churn probability in a field experiment. Three alternative courses of marketing action are applied: giving free incentives, organizing special customer events, obtaining feedback on customer satisfaction through questionnaires. The results of this field experiment show that profits can be doubled using our churn-prediction model. Moreover, profits vary enormously with respect to the selected retention action, indicating that a customer satisfaction questionnaire yields the best results, a phenomenon known in the psychological literature as the ‘mere-measurement effect’.
The pay-TV company offers premium-channel content on a variety of topics and live broadcasts using an encrypted signal. The programming of the channel is mainly based on recent movies that have not been broadcasted yet (on free TV), as well as main sports events. In addition, it also offers a variety of information, home-made programs and series. During the early 1990s, the pay-TV company grew expansively and obtained a considerable customer base by 1996. After a few years of stagnation, 2001 marked the start of a constant decrease in membership (see Fig. 1). This significant decrease in membership urged management to enhance the relationship-building process with their customers. This move was motivated by the specific nature of pay-TV broadcasting, which is characterized by fairly high fixed costs due to high infrastructure investments (settop box, proprietary technology) as well as high broadcasting costs.
This paper shows how a company operating on a subscription basis can remedy high attrition rates. We apply different binary classification techniques on this customer-churn case: logistic regression, Markov chains and random forests. We choose the most appropriate model for our case, and justify that choice.
Next to developing a churn-prediction model, a company should also test its attrition prevention strategy. Any such strategy has to be implemented through specific marketing actions. This section describes three such courses of action, which have been implemented in the present case-study, but which can be applied to subscription services in general. The action types are (1) giving free incentives (enhancing the service), (2) organizing special events to pamper customers, and (3) obtaining feedback on customer satisfaction through questionnaires. The results of these strategies will be reviewed in the context of the pay-TV market, but we feel confident that this approach can be generalized to any professional, membership, or subscription services business; e.g. internet service providers, utility providers, telephone services (mobile, long distance…), newspapers, journals and magazines, web-based information delivery, health care, financial services, computing services, insurance.
After pointing out the importance of managing customer churn, this paper describes the model-building process and evaluates different proposed models. Next, the attrition-prevention strategies and the field experiment are described (see Fig. 2). The discussion of the results includes a quantification of profits for the pay-TV company. In addition, the transfer of the benefits of the model to other subscription-dependent industries is highlighted and directions for further research are provided.
The pay-TV company encountered huge churn rates at the beginning of the twenty-first century. We have shown in this article, what to do when a company is in such a situation. We developed a churn-prediction model, and targeted potential churners with three different churn-prevention actions, using that model. The empirical results of three alternative courses of action reveal that all three are equally effective to reduce customer attrition: (1) giving free incentives (enhancing the service), (2) organizing special events to pamper customers and (3) obtaining feedback on customer satisfaction through questionnaires. Given the fact that the latter has the added benefit of increasing ‘knowledge’ about the individual customer, it seems to be the most attractive one. In summary, our results show that, using the full potential of our churn-prediction model and the incentives available, the pay-TV company’s profits of the churn prevention program would double when compared to its current model. We believe that a similar approach is applicable to most subscription or membership services.
