Over the years we have found that there are huge advantages to testing. However, we have also found that the process brings with it certain problems, some of which are not easily solved. One of these is explaining the value\cost of testing to clients, and deciding whether to confront the issue during the sales phase.
When we start to talk about testing we always seem to end up discussing integration testing. But is integration testing really our favourite superhero? Or can it become our worst enemy? In this post I'm going to try, using a practical example, to work out how far we can take integration testing and when it would be better to delegate testing to other parts of the application.