Modeling social coupon redemption decisions of consumers in food industry: A machine learning perspective

Abstract

Social couponing is a growing promotional phenomenon in the service industry. However, since the conversion rate of distributed coupons into coupons redeemed for purchase is relatively low, there is a need to understand the redemption decisions of consumers. Lower conversion rates lead businesses to lose both customers and profits. Previous studies have typically focused on social couponing from a business perspective, without exploring factors from the customer's end. The current study explores the factors influencing customers' decision to redeem coupons and highlights the interrelationships between the factors. Data were collected from 353 online customers on their redemption experiences during their food purchases. Structural equation modeling was performed to examine the significance of the factors and establish the predictability of customers' redemption decisions. We then explored different machine learners to identify the best-fitting models for customers' redemption decisions. Results showed that the prediction accuracy of the decision-tree-based models was the highest. These models delineate the role of influencers in various redemption aspects and validate the mediation effects of perceived risk, deal proneness, referral, and consumption frequency. The study also highlights future research areas in the social couponing domain.