The authors felt that certain changes were needed in Keaveney’s model to make it useful in understanding switching between banks. Two incidents were discarded and, of the six that remained, two were retitled. The reasons are offered below.
First, the authors suggest that the headings given to the third and fourth types of incident are changed to ‘service failures’ and ‘unacceptable behaviour, attitude or knowledge of staff respectively. The new titles, in the authors’ opinion, better reflect the various reasons that are categorised under the respective headings. A ‘service failure’ is a service failure irrespective of whether the failure did or did not take place with a core service. In a banking context, as most banks offer a broad range of services, a question which would need to be addressed is, should the word ‘core’ be retained, in ‘which services are core and which are non- core?’ By merely calling this type of switching incident ‘service failures’, there becomes no need to distinguish between core and non-core services. The title ‘unacceptable attitude, behaviour or knowledge of staff seems to illustrate better the reasons falling under this heading, as opposed to the use of the original title of ‘service encounter failure’.
Secondly, the authors considered the extent to which respondents would mention each of the eight incidents in a bank switching study. In regard to ‘unethical behaviour’, Singapore takes a strong stance against corruption (as evidenced, among other things, in a survey conducted during 1999 by the World Economic Forum: Singapore ranked equal fifth in the ‘Least corrupt’ index). In regard to ’employee responses to service failures’, as local banks employ and train customer service officers, it would be most unlikely that respondents would switch because of a response-related problem. If either ‘unethical behaviour’ or ’employee responses to service failures’ should be mentioned by one or more respondents, they would be recorded under a subset of the final incident, namely as ‘seldom mentioned incidents’.
This study, in summary, uses a six- incident model, with two incidents from the model by Keaveney (1995) being renamed. ‘Core service failures’ was renamed ‘service failures’ and ‘service encounter failures’ was renamed ‘unacceptable behaviour, attitude or knowledge of staff.