I think the first question is what data do you collect. For our testing area, we gather information on where the errors occurred, what test script, test wave, etc. This allows us to determine if we are finding more or less errrors in later test waves, are we finding reoccurring errors, what is our productivity in tested scripts, and even after the fact abends vs testing rigor.
We also look for leaning opportunities. There normally is a big time delay is restaging the test data for another round of testing, so we look opportunities to minimize outside set up time, vs inside set up, concurrent vs sequential work, etc.
.
