Internet banking has become widely available in Jamaica and yet there have been few studies to understand the characteristics of its users. For banks to improve their service and similar services it becomes imperative that they can justify the costs associated with these services. One of the ways these costs can be justified is if their customer base of internet banking users was to increase, this can be achieved if the characteristics (i.e. both demographic and behavioural) of the users can be identified. In this study, survey data of internet banking users and non-users was analyzed to develop predictive models which can be used to improve the uptake of this service. CRISP-DM methodology was used to improve the dependability and quality of the data mining results.