Book club system

h2.Second strategy
The second strategy considered was to take an existing club and split it so that new members would be handled by the new system while existing members would stay in the old one. Again, this would have allowed the project to focus on a core set of functionality. Unfortunately this strategy also turned out to be unfeasible. The killer detail this time was customer support. When customers call in they often do not know their customer id (they often do not know the name of their club!). With a club split across systems, customer support would have to look for the customer in both systems.

h2.Final strategy
The strategy that ultimately was chosen was to develop enough functionality to be able to migrate one of the smallest clubs. This was a larger [MRP] than the other strategies since the project now had to develop support for both new and existing customers as well as the migration code. This strategy worked well but there were some drawbacks. Customer support still had to cope with two systems but now they knew which system to use as long as the customer remembered which club they belonged to. Some customers were split across both systems since they belonged to more than one club. Finding and rejecting bad customers was complicated by the fact that records of bad customers resided in two systems.