The theory-practice research gains from big data: evidence from hospitality loyalty programs

Abstract

Purpose The hospitality industry values segmentation and loyalty programs (LPs), but there is limited research on new methods for segmenting loyalty program members, so managers often rely on conventional techniques. This study aims to use big data-driven segmentation methods to cluster customers and provide a new solution for customer segmentation in hotel LPs. Design/methodology/approach Using the k-means algorithm, this study examined 498,655 profiles of guests enrolled in a multinational hotel chain’s loyalty program. The objective was to cluster guests according to their consumption behavior and monetary value and compare data-driven segments based on brand preferences, demographic data and monetary value with loyalty program tiers. Findings This study shows that current tier-based LPs lack features to improve customer segmentation, and some high-tier members generate less revenue than low-tier members. Therefore, more attention should be given to truly valuable customers. Practical implications Hotels can segment LP members to develop targeted campaigns and uncover new insights. This will help to transform LPs to make them more valuable and profitable and use differentiated rewards and strategies. Originality/value As not all guests or hotel brands benefit equally from LPs, additional segmentation is required to suit varying guest behaviors. Hotel managers can use data mining techniques to develop more efficient and valuable LPs with personalized strategies and rewards.