Abstract
Research suggests that the greater service failure severity the greater the recovery performance necessary to convert the customers' dissatisfaction into satisfaction and the more difficult it should be for the service provider to achieve full recovery (Magnini, Ford, Markowski, and Honeycutt 2007; Mattila 1999; McCollough, Berry, and Yadav 2000; Smith and Bolton 1998; Sundaram, Jurowski, and Webster 1997; Weim, Beatty, and Jones 2004). This research investigates the recovery paradox, the proposition that superior recovery can leave the customer as satisfied, if not more satisfied, than if nothing had gone wrong (McCollough and Bharadwaj 1992) by examining the impact of service failure severity and the recovery performance on post-recovery satisfaction. This research answers the call for future research investigating service failure severity (Matos, Henrique, and Rossi (2007) extending previous research on post-recovery satisfaction and the recovery paradox. Findings show that for a recovery paradox to emerge the service failure severity must be very modest and the recovery effort superior. Indeed, the relative low level of harm caused by the failure and the relatively high recovery performance necessary is surprising and indicates that the recovery paradox may be a rather limited phenomenon. [PUBLICATION ABSTRACT]