Predicting Customer Churn for Insurance Data

Scriney, Michael; Nie, Dongyun; Roantree, Mark

doi:10.1007/978-3-030-59065-9_21

Cited by 14 publications

(12 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Customer churn prediction modeling has often been the focus of researchers, as evidenced by numerous studies published on this topic. Particularly well-explored are the contractual business settings in the B2C domain, such as those commonly encountered in the telecommunications [19][20][21], banking [22,23], and insurance [24,25] sectors, where customers at risk of terminating or not renewing their contracts are identified and targeted with retention campaigns in efforts to persuade them otherwise. Non-contractual settings have also often been studied, where efforts have been put towards predicting which retail customers are least likely to make a purchase in the future [26,27], which users are at most risk to stop playing mobile games [6], or which passengers are not planning to use a particular airline for their future flights [28].…”

Section: Literature Reviewmentioning

confidence: 99%

Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

et al. 2022

View full text Add to dashboard Cite

Customer churn is a problem virtually all companies face, and the ability to predict it reliably can be a cornerstone for successful retention campaigns. In this study, we propose an approach to customer churn prediction in non-contractual B2B settings that relies exclusively on invoice-level data for feature engineering and uses multi-slicing to maximally utilize available data. We cast churn as a binary classification problem and assess the ability of three established classifiers to predict it when using different churn definitions. We also compare classifier performance when different amounts of historical data are used for feature engineering. The results indicate that robust models for different churn definitions can be derived by using invoice-level data alone and that using more historical data for creating some of the features tends to lead to better performing models for some classifiers. We also confirm that the multi-slicing approach to dataset creation yields better performing models compared to the traditionally used single-slicing approach.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

et al. 2022

View full text Add to dashboard Cite

show abstract

“…All methods were fitted to LGPIF data from years 2006 to 2012, and out-ofsample validation was performed on data in year 2013. In terms of assessing performance, a widely used evaluation metric in traditional customer churn analysis is error rate = (1 − accuracy), which is defined as the percentage of incorrectly classified observations in the test set (see for example Bolancé et al, 2016;Loisel et al, 2019;Scriney et al, 2020). However, a major disadvantage of error rate is that it is not well suited to imbalanced data with rare events (with imbalanced classes, it is easy to get a low error rate without actually making useful predictions, Morrison, 1969), and this is the case here with the LGPIF data, as exemplified in Table 1.…”

Section: Assessing Out-of-sample Performancementioning

confidence: 99%

“…We ensured the numbers of transitions with different states of destination (full-coverage, partial-coverage, and churn) were equal by adding additional samples to the minority transitions with replacement. In contrast to oversampling, undersampling is also widely used in customer churn analysis when a data set includes a large number of records; see Scriney et al (2020). However, as discussed in Section 4, the LGPIF data do not contain enough time points or observations for this to be relevant.…”

Section: Aucmentioning

confidence: 99%

Multi-State Modelling of Customer Churn

Dong

Frees²,

Huang³

et al. 2022

ASTIN Bull.

View full text Add to dashboard Cite

Customer churn, which insurance companies use to describe the non-renewal of existing customers, is a widespread and expensive problem in general insurance, particularly because contracts are usually short-term and are renewed periodically. Traditionally, customer churn analyses have employed models which utilise only a binary outcome (churn or not churn) in one period. However, real business relationships are multi-period, and policyholders may reside and transition between a wider range of states beyond that of the simply churn/not churn throughout this relationship. To better encapsulate the richness of policyholder behaviours through time, we propose multi-state customer churn analysis, which aims to model behaviour over a larger number of states (defined by different combinations of insurance coverage taken) and across multiple periods (thereby making use of readily available longitudinal data). Using multinomial logistic regression (MLR) with a second-order Markov assumption, we demonstrate how multi-state customer churn analysis offers deeper insights into how a policyholder’s transition history is associated with their decision making, whether that be to retain the current set of policies, churn, or add/drop a coverage. Applying this model to commercial insurance data from the Wisconsin Local Government Property Insurance Fund, we illustrate how transition probabilities between states are affected by differing sets of explanatory variables and that a multi-state analysis can potentially offer stronger predictive performance and more accurate calculations of customer lifetime value (say), compared to the traditional customer churn analysis techniques.

show abstract

“…While this reflects a real and common problem for organizations, the two most popular variables for CLV calculations were missing: acquisition and retention. Thus, the imputation of these values is part of our future research and will be published in Scriney et al (2020).…”

Section: Conclusion and Future Plansmentioning

confidence: 99%

A methodology for classification and validation of customer datasets

Nie

Cappellari

Roantree

2020

JBIM

Self Cite

View full text Add to dashboard Cite

Purpose The purpose of this paper is to develop a method to classify customers according to their value to an organization. This process is complicated by the disconnected nature of a customer record in an industry such as insurance. With large numbers of customers, it is of significant benefit to managers and company analysts to create a broad classification for all customers. Design/methodology/approach The initial step is to construct a full customer history and extract a feature set suited to customer lifetime value calculations. This feature set must then be validated to determine its ability to classify customers in broad terms. Findings The method successfully classifies customer data sets with an accuracy of 90%. This study also discovered that by examining the average value for key variables in each customer segment, an algorithm can label the group of clusters with an accuracy of 99.3%. Research limitations/implications Working with a real-world data set, it is always the case that some features are unavailable as they were never recorded. This can impair the algorithm’s ability to make good classifications in all cases. Originality/value This study believes that this research makes a novel contribution as it automates the classification of customers but in addition, the approach provides a high-level classification result (recall and precision identify the best cluster configuration) and detailed insights into how each customer is classified by two validation metrics. This supports managers in terms of market spend on new and existing customers.

show abstract

Predicting Customer Churn for Insurance Data

Cited by 14 publications

References 16 publications

Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

Multi-State Modelling of Customer Churn

A methodology for classification and validation of customer datasets

Contact Info

Product

Resources

About