Amy Smith

Who they are

This page records the public professional record only. Nothing adverse is recorded and no motive is attributed.

Her affiliation and wider career were not verified for this page, so none is asserted. The 1999 paper also circulated earlier as a Marketing Science Institute working paper under the same title. Ruth N. Bolton, one of her co-authors, is a separate researcher from Lisa Bolton, whose work on price fairness appears elsewhere on this site.

Key papers

Smith, A. K., Bolton, R. N., & Wagner, J. (1999), 'A Model of Customer Satisfaction with Service Encounters Involving Failure and Recovery', Journal of Marketing Research, 36(3), 356-372, doi:10.1177/002224379903600305. Read at the level of the published abstract and the working paper summary.

Her wider publication record was not surveyed for this page.

Key ideas

The paper starts from an observation: customers often react strongly to service failures, so a recovery has to be equally strong and effective. The question is what strong means.

The model answers it with an exchange framework. It integrates the consumer satisfaction and social justice literatures using three sets of principles: resource exchange, in which a failure takes something and a recovery gives something back; mental accounting, in which customers keep separate accounts for different kinds of gain and loss; and prospect theory, in which losses weigh more than equivalent gains.

The test was a mixed-design experiment delivered as a survey, in which customers evaluated failure and recovery scenarios and answered questions about an organisation they had recently used. It ran in two settings, restaurants and hotels. The results showed that customers prefer to receive recovery resources that match the type of failure they experience, in amounts commensurate with the magnitude of the failure. The authors framed that as guidance for managers on achieving the proper fit between a failure and the recovery effort.

The fit idea was later tested at scale. Holger Roschk and Katja Gelbrich's 2014 meta-analysis and experiment found the recovery effect generally strongest when compensation was a resource similar to the failure, immediate money for a monetary loss, an exchange for a flawed product, reperformance for a failed service, with the exception that a lack of attention could be repaired by other kinds of compensation as well. The amount side was tested by Gelbrich and colleagues in 2015, who found satisfaction rose fastest between about 60 and 120 per cent of the loss when customers had rejected the flawed service.

The paper is one of the outcome-side foundations of the article What a Customer Apology Must Contain.

Limits on the 1999 model

The 1999 model was tested with scenarios. Customers answered questions about failures that were described to them, in two service settings, restaurants and hotels, so the result cannot carry the anger of a failure that really happened. Its claim is about preference: customers favoured recovery resources that matched the failure, in proportion to how bad it was. Preference is not behaviour, and the later work by Roschk and Gelbrich in 2014, which pooled studies and ran a fresh experiment, is the better evidence for how strongly the matching effect shows up. The abstract is also all that was read here, so details of the design beyond what it states are not reported. What the paper gave the field is the idea of fit and a vocabulary, resource exchange, mental accounting and prospect theory, in which to ask whether a recovery returned the kind of thing the failure took.

---