A Meta-Analysis on Loyalty Program Membership Effects –The Influence from Firm, Industrial, and National Levels

Abstract

Loyalty programs are a common customer relationship management tool that has been adopted in various industries. Despite their prevalence, research on loyalty programs find inconsistent results on loyalty program effectiveness in terms of magnitude and direction. To clarify the effects of loyalty programs, the first aim of this thesis is to investigate whether loyalty program membership has an impact on a range of customer responses. A meta-analysis is used to solve this research question. In total, 432 effect sizes on the relationship between loyalty program membership and customer responses from 81 independent samples were collected. The average corrected sample size-weighted correlations show loyalty program membership generally has a positive yet small effect (r < .30) on 17 customer response outcomes. The results from the first meta-analysis also show substantial heterogeneity which is caused by between-study differences other than random sampling errors. Therefore, the next research question is to identify sources of heterogeneity in loyalty program effects, i.e. a moderator analysis on the underlying factors that influence the relationship between loyalty programs and customer response variables. Drawing on the existing research on loyalty programs, three levels of potential moderators were proposed. At the firm level, program structures and firm size were assessed. At the industrial level, the model incorporates product characteristics and market concentration. At the national level, Hofstede’s five cultural dimensions and national economic factors were tested. A number of variables of study characteristics were included to control for different study designs. To further explore loyalty program effects in complex situations, this study tests the interactions between national culture and product characteristics. This thesis provides an overview of the current research on loyalty programs by quantitatively integrate existing research results. It identifies the strength of loyalty program effects, which are generally weak. Therefore, managers should carefully evaluate the use of loyalty programs for their businesses, given the high initial investment of launching such a loyalty initiative. More importantly, this thesis assesses three levels of moderators that might influence the strength of the loyalty program effects within a single framework.Managers should take into account of these factors examined when evaluating and designing their loyalty program strategies to optimise the output of loyalty programs.