Guest preferences for service recovery procedures: conjoint analysis

Abstract

PurposeThe ongoing challenge for hospitality firms is not to avoid service failures completely but rather to find effective ways of recovering from them. The purpose of this paper is to adopt conjoint analysis to examine guest preferences for service recovery actions that are taken simultaneously and the effect of such actions in the context of hotels in order to seek answers to the following research questions: what type of compensation bundle is most preferred by guests? and How effective are compensations in terms of recovery satisfaction, especially with regard to subsequent behavioral intentions?Design/methodology/approachIn order to study guest preferences for compensation, choice-based conjoint (CBC) experiments were used. The CBC survey method involves presenting respondents with several hypothetical scenarios that include numerous combinations of compensation attributes. The survey sample was recruited through web-based panels run by a marketing firm. The online survey was designed and built in the Sawtooth Software platform. Conjoint analysis was carried out.FindingsThe conjoint analysis results show that among the different compensation attributes, consumers preferred discounts the most, followed by corrective actions and loyalty points. Among the discount levels, consumers favored “free one night certificate for a future stay” over “100 percent off one night’s room bill.” Regarding loyalty points, consumers desire more points than fewer points. In terms of correction levels, consumers appear to desire an “upgrade to a suite” over just moving to another room or having the original room cleaned. Moreover, consumers prefer that service recovery be handled by upper-level management.Research limitations/implicationsSeveral limitations should be discussed. For example, the type of service failure studied was limited to a booked room not being clean upon check-in. Different types and magnitudes of service failures may result in a different set of preferences for compensation bundles. In addition, the compensation attributes and levels were limited in scope even though they were obtained from interviews with managers at midsize hotels. Other attributes and attribute levels can be included in future studies.Originality/valueAlthough an abundant number of studies have been done regarding service failure and recovery, the extent to which service recovery needs to be carried out before it translates into actual recovery remains unanswered. Using conjoint analysis, the study identified exactly which bundle of compensation items was most preferred by guests.