Profit-Based Model Selection for Customer Retention Using Individual Customer Lifetime Values

Abstract

The goal of customer retention campaigns, by design, is to add value and enhance the operational efficiency of businesses. For organizations that strive to retain their customers in saturated, and sometimes fast moving, markets such as the telecommunication and banking industries, implementing customer churn prediction models that perform well and in accordance with the business goals is vital. The expected maximum profit (EMP) measure is tailored toward this problem by taking into account the costs and benefits of a retention campaign and estimating its worth for the organization. Unfortunately, the measure assumes fixed and equal customer lifetime value (CLV) for all customers, which has been shown to not correspond well with reality. In this article, we extend the EMP measure to take into account the variability in the lifetime values of customers, thereby basing it on individual characteristics. We demonstrate how to incorporate the heterogeneity of CLVs when CLVs are known, when their prior distribution is known, and when neither is known. By taking into account individual CLVs, our proposed approach of measuring model performance gives novel insights when deciding on a customer retention campaign. The method is dependent on the characteristics of the customer base as is compliant with modern business analytics and accommodates the data-driven culture that has manifested itself within organizations.

Profit-driven churn prediction for the mutual fund industry: A multisegment approach

“…Several studies have emphasized the importance of using individual CLVs rather than a single value (generally the average CLV); see [2,3,36] . Although the framework by Verbeke et al [44] allows taking on a distribution for the CLV, this has not been done empirically in most EMPC studies.…”

“…Following the framework designed by Safari et al [40] for the banking industry, we consider T = 3 years, r equals the inflation rate, and the direct costs about the accounts Cost j,t to be negligible. Parameter r t is the retention rate for a given period, i.e., the probability of a customer staying in the company at time t [36] .…”

“…Although the importance of considering individual CLVs has been acknowledged in most of the studies in the area (see e.g. [36,44,47] ), the first profit metrics developed did not account for heterogeneous CLVs, they averaged them. Recently, Óskarsdóttir et al [36] computed individual CLVs but considered a single decision threshold for a given classifier.…”

“…[36,44,47] ), the first profit metrics developed did not account for heterogeneous CLVs, they averaged them. Recently, Óskarsdóttir et al [36] computed individual CLVs but considered a single decision threshold for a given classifier. This approach does not allow to distinguish between the most valuable customers and the rest, ending up aggregating customers in such a way that the use of a single average CLV is almost equivalent to the use of individual CLVs.…”

Comparative Analysis of Selected Probabilistic Customer Lifetime Value Models in Online Shopping

“…The main applications of CLV target to business-to-consumer (B2C) context while ignoring business-to-business applications of customer asset management (Nenonen & Storbacka, 2016). The relevance of CLV applications is primarily utilized at the customer level, which affects customer loyalty (Qi, Qu, Zhou, & Li, 2015) and customer behaviour impacting retention (Abdolvand, Baradaran, & Albadvi, 2015;Óskarsdóttir, Baesens, & Vanthienen, 2018). According to the results of (Kumar & Pansari, 2016) the country's economy has a direct impact on the frequency of purchases and the margins of purchases and attention should be paid to cultural dimensions and economic conditions due to their different impact on CLV.…”