Mass customization for financial services: an empirical study of adoption and usage behavior

Purpose – This paper aims to report findings from an exploratory empirical study focusing on an application of mass customization in financial services. Based on the study of configurations and usage data, the authors evaluate a series of hypotheses relating to the interplay of adoption and usage by customers. Design/methodology/approach – The study is based on quantitative analysis of data from a Turkish bank which offers customizable credit cards, encompassing both configurations as well as credit card usage. Findings – The results confirm that trial-and-error learning will not end with product definition, but will continue afterwards and lead to changes in customization. Especially active usage length shows a significant positive effect on the number of changes. The effect of base category usage could only partly be confirmed for changes, but was significant for adoption. It was also found that a series of smaller changes in a limited number of attributes has a higher likelihood than a smaller number of changes in a large number of aspects. Research limitations/implications – The study uses data from a single financial service provider, from a specific country. In addition, anonymized data on adoption and usage were used, thus demographic data as well as subjective measures from customers were not available. Practical implications – The results highlight the importance of specifying the correct solution space, as the authors could at least partially confirm the negative effect of both a large number of options, as well as basing on alternatives rather than attributes on several levels. Although overall mass customization seems less interesting than traditional credit cards, the authors discuss several positive implications for financial sector companies from offering this option. Originality/value – The paper extends current literature in focusing for the first time on mass customization for financial services. In addition, this is the first study using longitudinal data on adoption and modification of mass-customized solutions to analyze the long-term behavior of usage.