James G. Maxham III and Richard G. Netemeyer

The Economist (2000) reported that customer complaints are rising sharply. While retailers cannot eliminate complaints, they can learn to effectively respond to them. Some researchers suggest that a retailer's response to failures can either reinforce customer relationships or exacerbate the failure. The existing literature is mostly comprised of laboratory or field experiments based on hypothetical scenarios, or, cross-sectional studies in which respondents were asked to "think back" to some past failure. Though these studies have contributed to our understanding of complaint handling and service recovery, field studies are needed that capture customer perceptions of failures and recoveries as they unfold over time. Also, given the importance of relationship marketing in ongoing service industries, studies are needed to determine if satisfaction gains realized by offering fair redress in the service recovery transaction affect overall firm satisfaction.

We tested a perceived justice and satisfaction-based model of the service recovery process as it takes place over time. We conducted two field studies examining the effects of justice on satisfaction with recovery, overall firm satisfaction, purchase intent, and the likelihood of spreading positive word-of-mouth (i.e., WOM invent). We also attempted to reduce common recall biases in recovery research by capturing customer perceptions as they most likely form over time. Study One focuses on 692 customers who actively complained about the service at one of 116 bank branches of an industry-leading bank. Study Two focuses on 339 homebuyers that complained to their homebuilder at one of 215 communities across 12 states. Three surveys were administered across time: one at the time of complaint; one following the recovery efforts; and one two weeks following the recovery efforts.

While our research suggests that retailers should strive to offer customers distributive, procedural, and interactional justice, it also suggests directions for positively affecting different types of customer intentions. Retailers offering refunds and discounts (i.e., distributive justice) following product or service failures can likely increase satisfaction with recoveries and indirectly affect WOM intent. However, managers should not discount the importance of procedural
and interactional justice. Beyond their direct effects on overall firm satisfaction, procedural and interactional justice
also directly affected WOM intent and purchase intent (i.e., not totally mediated by satisfaction). Notwithstanding, it
appears that distributive justice is a key antecedent of satisfaction with recovery and WOM intent. As such, in durable good industries where repurchases are less frequent and recommendations are vital, it may prove beneficial for retail
managers to offer proportionately more distributive justice.

Our studies also suggest that retailers offering procedural and interactional justice following failures may increase overall firm satisfaction and indirectly affect purchase intent. As such, it appears that fair policies and procedures, as well as courteous employee interactions, influence more enduring customer perceptions of overall firm satisfaction and purchase intent. Although retailers should strive to provide distributive justice, they may see the greatest impact on purchase intent by concentrating on procedural and interactional justice. Thus, in ongoing service relationships where long-terms loyalty is a cornerstone of success, it may prove beneficial for managers to focus more on procedural and interactional justice in recovery strategies.

We also offer implications for service and durable good failures. Our results suggest that distributive justice is more pronounced in forming satisfaction with recovery among durable good complainants than service complainants, and interactional justice is more influential in forming satisfaction with recovery among service complainants than durable good complainants. This suggests that durable good complainants are relatively more concerned with redress outcomes (i.e., distributive justice) than with customer service interactions. As such, durable good managers may benefit from investing proportionately more dollars in either "fixing" the failure or offering compensation than in customer service training. Finally, our results suggest that procedural and interactional justice are particularly pronounced in service industries. Service policies and employee-customer interactions are often seen as an integral part of the service offering (Schneider et al., 1998). When forming perceptions of satisfaction with recovery, service complainants seem to weigh policies and buyer-seller personal interactions more heavily than durable good complainants weigh such process variables. Thus, for pure services, a heavier focus on procedural and interactional justice may be a viable managerial strategy.