role. A key decision is then whether to try to fit the bank to the system or the other way round.Most banks now opt for the former approach, at least in theory – sometimes, as the implementation begins, such ideals are lost as end- users seek a system that replicates what they currently do. The theory was eloquently put by UnionBank of the Philippines' senior vice president, Ramon Duarte: 'We are a bank, not a software company, and we are not of a size that we can dedicate a lot of resources and people so to a degree we decided to trust the vendors – they are the process scientists.'
Often, while a large portion of the selection time is centred around functional specifications, the final decision comes down to other factors. If a bank is fairly standard in its operations, it might well find that there is little difference between the functionality of its shortlisted packages. In this case, aspects that come into play include commercial terms, reference site feedback, delivery track-record, and local support.
The advantages of working with a smaller supplier are cited by a number of banks. Rabobank, for instance, felt this was a benefit with its projects with both Callataÿ &Wouters and Calypso (the latter supplier was considerably smaller than it is today at the time of the Dutch bank's project).
References can work for and against a supplier. For instance, Industrial Development &Workers Bank of Egypt (IDWBE) took up references from its four shortlisted vendors. But, said board member, DrMoetassem Bellah Kaddah, 'not all of them were successful references. On the contrary, some of them gave references that intimidated us.'
