The paper's aim is to test the existence of the service recovery paradox.
To date, much of the literature exploring the service recovery paradox has generated mixed results. The paper argues that a service recovery paradox is a rare event, which makes its measurement difficult, since the “treatment group” sample size is usually too small to produce significant results. For that reason, the existence of the service recovery paradox in a banking context with more than 11,000 customer interviews based on actual customer encounters is tested.
Overall, the survey findings support the argument that a service recovery paradox is a rare event, and the hypothesized mean differences are, albeit significant, not very large, which diminishes their managerial relevance to some degree.
Because of the required extremely large sample size, no multi‐item measures were collected. Furthermore, privacy concerns restricted us from a longitudinal study and from linking the survey results to behavioural data. Both limitations are inherent in the chosen setting.
While a service failure offers an opportunity to create an excellent recovery, the likelihood of a service paradox is very low. The implications of verifying a service recovery paradox do not suggest that ineffective service followed by an outstanding service recovery is a viable strategy.
To the authors' knowledge, this is the first empirical study testing not only the existence of the service recovery paradox, but also exploring its magnitude and frequency. This is crucial because the paradox is a very rare event, which, in turn, limits its managerial relevance.
