What makes you (s)tick?

What makes you (s)tick?

An empirical study on factors inducing customer switching

behavior among Dutch Health Insurance Customers

Diederick Dorenbos

MSc Marketing Intelligence & Marketing Management Thesis defense

Table of Contents

1. Relevancy

2. Conceptual model

3. Data Collection

4. Findings & Discussion

5. Implications

Relevancy

The Customer of today Value Concious Consumer

Intolerant to low quality (Digitally) conncected to each

Relevancy

Why a marketing manager should know?

More switching and less loyalty

Increasingly quality and price conscious

consumers

Firm value can reduce billions of dollars due to high

churn rates1

Focus on Long-Term customer

Relationships

Relevancy

For marketing managers to allocate resources to retention efforts, two elements of customer switching behaviour are key.

Relevancy

Don’t we know, already?

Identified limitations of predictors in prior research. Customer

Research Questions

“What are the drivers of customer switching behavior at the customer level?”

Research Question 1: What is the

effect of customer dissatisfaction on switching?

Research Question 2: What is the effect

of customer engagement on customer switching behavior at the customer level?

Research Question 3: What role does

the intention to switch play in actual switching behavior at a customer level?

Research Question 4: Does the

Conceptual Model

Data Collection

Methodology

Analyses

› Conventional techniques and machine learning techniques

› Is intent to churn a separable construct? › Is customer behaviour in line with their own stated aspects of switching?

Analysis 1: Binomial Logistic Regression Analysis 2: Out-Of-Sample Predictions

Analysis 3: Mediation Analysis Analysis 4: Churn in Retrospect

› Binary Choice Model

› Signs/Marginal Effects/Odds-ratio

Findings

What is the likelihood to churn affected by?

▪ Lower prices at

competitor increase customer churn

likelihood.

▪ High perceived price at current insurer does not. ▪ Overall, no concluding evidence. Customer dissatisfaction (Economic) Customer dissatisfaction (Service)

▪ Dissatisfaction with service level leads to increased customer churn.

▪ Dissatisfaction with

customer with insurance contract does not.

Findings

What is the likelihood to churn affected by?

▪ Did not yield significant effect on churn likelihood Customer Engagement Customer Characteristics ▪ Being in a higher age

Findings

Intentions to churn by a customer

▪ Intention to churn is a strong predictor of actual churn.

▪ But is it a mediator?

▪ No convincing mediating role with actual churn.

Churn Intent Churn Intent

▪ Treat churn and churn intent as separate construct.

▪ Dissatisfaction with offered

service and price level of health insurer increases intent to churn. ▪ Customer engagement

Findings

What prediction technique should a marketing manager use?

▪ Commonly used Logistic Regression performed best for TDL

▪ Hit-rate was highest for Support Vector Machine (SVM)

▪ Surprisingly, more complex (ML)

methods may not always perform best

Findings

Churn in Retrospect Churn in Retrospect

▪ Why did you switch or stay? ▪ Compare with predictors of

churn

▪ Pricing (54%) ▪ Service (19.8%)

▪ Customer Engagement (3.8%)

Churn in Retrospect ▪ Pricing most substantial

Key-takeaways

▪ No concluding evidence of increased churn for economically dissatisfied customers. ▪ No concluding evidence of increased churn for service-related dissatisfied customers. ▪ Churn intent can be predicted by customer Engagement

▪ In this research context, engagement is indicative of increased churn intent ▪ The intent to churn acts as separate construct from actual churn with different

predictor variable significances.

Key-takeaways

▪ Economic and service related determinants of satisfaction are not certain predictors of churn in the context of dissatisfied customers.

▪ Customer churn and customer churn intent should be treated as separate topics of interest.

▪ The older segment of customers is less likely to churn than younger segments ▪ Cost-effective retention efforts should differentiate.

▪ In retrospect, customers state the importance of engagement on their decision making

Limitations

▪ Context of Health Insurance > May not be generalizable ▪ Customer Engagement

▪ Effect of dissatisfaction as to price/service may differ across industries ▪ Limited to contractual churn setting

▪ Observations could have been larger across the industries, yet many incomplete cases