The last few years have seen many changes in the banking world, but we’ve recently experienced a new effect of the financial crisis; starting a project before the user community is in place. Not only did this regulatory-driven project have continuously changing requirements and parameters, but the extreme circumstances meant that various aspects of the programme were running in parallel and to a delivery schedule with zero contingency (of course).

Traditionally, users define the business requirements at the outset and the project team works through the four stage process, analysis, design, build and test, before returning the new system to the users with complete knowledge transfer.  In this case, the critical state of affairs caused by the banking crisis created a project where the requirements were in a continuous state of change or refinement and design, development and testing were all happening in parallel. In the end the team had to devise a closed scope design in order to enable an iterative approach to development and make some progress.

When the user community were eventually hired, the project moved into a concurrent testing phase, which resulted in a raft of change requests. The advantage of the team’s parallel, iterative and integrated approach however, was to enable continuous feedback between the sub-teams during analysis and design. Whilst unorthodox, and we wouldn’t usually set out to do it this way, it did mean that we were conducting user acceptance testing and resolving issues in real time.

All in all, this was a particularly demanding project for the combined client and outsourced team. I wouldn’t choose to start a project without users; I wouldn’t choose to have such an extremely challenging deadline, but we have learnt that it’s possible. An unexpected effect of the financial crisis is that we developed a new way of working.

Graham Underwood, Managing Director, GFT UK.