Purpose – research has shown that more than half of attempted recovery efforts only reinforce dissatisfaction, producing a “double deviation” effect. Surprisingly, these double deviation effects have received little attention in service marketing literature. To fill this gap, this paper aims to develop and empirically test a model of how customers form satisfaction judgments in double deviation scenarios. The paper seeks to propose that emotions have a distinct and separate influence from perceived justice in explaining satisfaction with failed recovery.design/methodology/approach – the paper employs the critical incident technique to obtain data from banking customers and apply latent variable path analysis to test the proposed model.findings – the results of the study support the model and highlight the important role of specific recovery-related emotions in double deviation contexts.research limitations/implications – future research should try to determine whether different specific negative emotions and/or the interactional and procedural components of justice affect post-recovery judgments in double deviation scenarios.practical implications – the results show that specific emotions such as anger play an important role in explaining satisfaction with service recovery. The paper proposes that in future, customer satisfaction surveys could include items measuring specific emotions. This could increase their efficiency as managerial tools.originality/value – to the authors' knowledge, this has been the first attempt to model the effect of specific emotions triggered by the service recovery on satisfaction with service recovery and to empirically test a model of satisfaction with service recovery in double deviation scenarios. Furthermore, this study is based on the analysis of real service failures and recovery strategies.