We test a Life & Pensions Admin system and as such, we need to load policies onto the system and prove multiple tests from loading those policies at various stages of the process. These tests will generally prove different requirements from different workstreams/projects.

Loading a New Pension may have separate tests, such as; the right funds can be selected for investment, a new commission style can be selected and the right output can be issued to the client.

We're struggling to find a way to do this efficiently. As technically the only parts we are actually testing are the individual items and not the fact you can load the policy in general (we do that bit in Acceptance Test), we need to be able to include the tests as steps in the script, but not actually be part of it (so that if a non-test step fails, it doesn't fail the tests and if a test fails, the overall process can still pass.)

Currently we are having to write 1 script for the entire process (we call this a navigational script and don't link it to any requirements), then have separate test cases for each test which we link the requirement coverage to). We then reference each test case as a basic step in the navigational script and pull all of them together in a test set. This approach is far from ideal as running the test set means a lot of swapping over between the tests and the navigational script.

We've tried with "Call to Test", but this seems to work in exactly the opposite way to how we need it to. By this I mean that it seems we could call navigational scripts into a test, obviously thats no use to us as we need to load the whole process so we can re-use the policies for further tests later on.

Is there a way to use Call to Test differently. We've had very little training on TD other than the basics.