The Effects of Severity of Failure and Customer Loyalty on Service Recovery Strategies

Service failures do not need to result in permanent negative consequences as long as effective recovery activities are undertaken. Unfortunately, existing research has been limited in providing information to support prescriptive approaches for applying specific service recovery techniques. By using data from a large sample (n = 861) of service failure incidents and employing the use of hierarchical and non‐hierarchical cluster analysis, this exploratory study creates and analyzes empirical types of service failures. The derived failure types, or common situations faced by service providers, focus on customer loyalty and the severity of the failure, and may be visualized in a two‐by‐two matrix. Regression analysis is then used to demonstrate how effective recovery strategies and supporting activities should vary, based on the location of the failure within the matrix. The approach and results offer important implications for strategy and service support activities as well as a foundation for systematizing service recovery efforts.