Reference Dependent Targeting: Evidence From The Casino Industry
Loss aversion, a fundamental aspect of consumer behavior and marketing, has traditionally been regarded as a constant and relatively stable trait across individuals. However, recent research utilizing field surveys has demonstrated the heterogeneous nature of loss aversion and thus individual's sensitivities to reference points. Our study incorporates this growing field by exploring reference-dependent sensitivities at the individual and segment level using rich, individual level CRM data from a large US casino, and incorporating these insights to enhance CRM practice. We employ a hierarchical Bayesian and latent class modeling approach to analyze a unique panel dataset of over 12,000 slot machine players encompassing their playing activities and outcomes, revealing heterogeneity in reference-dependence sensitivities and the impact of previous gaming results on subsequent actions. Our flexible model accounts for various forms of reference construction (i.e., based on actual or theoretical outcomes), demonstrating that gamblers construct references distinctively, and different constructions yield varying degrees of sensitivities. Additionally, we reveal that reference-dependent sensitivities interact with marketing strategies and shape subsequent behaviors. From a managerial perspective, our findings indicate that incorporating heterogeneous reference-dependent sensitivities into a casino's CRM strategy (even at a segment level) can increase targeting efficiency by 11.7% compared to models that do not incorporate reference dependent sensitivity, and vastly outperform current industry practice.
casino, customer relationship management, loss aversion
SMU Cox: Marketing (Topic)