Inspection does not improve the quality, nor guarantee quality. Inspection is too late. The quality, good or bad, is already in the product. As Harold F. Dodge said, “You can not inspect quality into a product.”

— Out of the Crisis, page 29

I was asked this question a lot. I assumed most folks who asked me that question knew that finding and fixing a defect earlier costs less which they did. However, there was this belief that executing an end-to-end test was more efficient and so overall it would cost less to run those. Other times, I was told they tried BDD or trying to test earlier but it slowed them down or was unsustainable. Another challenge they saw was how can you even test something if it’s not complete. I decided to leave the QA team when I finally had data from my QA team and the COBOL development team to back-up what I already believed.

  1. Can QA tests be run against an incomplete system?
  2. Can all the end-to-end tests be run earlier?
  3. Does running the QA team tests during coding slow developers down?
  4. Does it reduce how much time QA has to execute tests?
  5. Does it matter if the tests actually drive the development?
  6. At what point should developers be involved?