Start Date

10-12-2017 12:00 AM

Description

While digitalization is prevalent in loyalty programs, little remains known regarding the relationship between mobile loyalty app usage and reward redemption behavior. Using a large panel dataset consisting of 201 million transactions made by 4.9 million customers over a period of two years, we implement both the hidden Markov model and data mining approach to grasp an understanding of the relationship. The hidden Markov model is developed to capture the dynamics in latent state transitions of customers depending on the type of loyalty program used and the predictive model is used to assess the informative value embedded in customers’ reward redemption behaviors before and after mobile app adoption. Our findings lend support to the prominent value of point redemption behavior in mobile-driven loyalty schemes. Our results also indicate a volatile state transition behavior of mobile app users highlighting potential pitfalls in an absence of appropriate marketing actions.

Share

COinS
 
Dec 10th, 12:00 AM

Loyalty Analytics: Predicting Customer Behavior Using Reward Redemption Patterns under Mobile-App Reward Scheme

While digitalization is prevalent in loyalty programs, little remains known regarding the relationship between mobile loyalty app usage and reward redemption behavior. Using a large panel dataset consisting of 201 million transactions made by 4.9 million customers over a period of two years, we implement both the hidden Markov model and data mining approach to grasp an understanding of the relationship. The hidden Markov model is developed to capture the dynamics in latent state transitions of customers depending on the type of loyalty program used and the predictive model is used to assess the informative value embedded in customers’ reward redemption behaviors before and after mobile app adoption. Our findings lend support to the prominent value of point redemption behavior in mobile-driven loyalty schemes. Our results also indicate a volatile state transition behavior of mobile app users highlighting potential pitfalls in an absence of appropriate marketing actions.