Category: Software Testing Advice

Good. You have noticed that test cases are not a unit of confidence.

If your test team is judged by how many defects it raises, congratulations: you have invented a system that rewards people for finding as many defects as possible.

It could mean the software is excellent. It could mean the testers are excellent and defects are being prevented or found earlier. It could also mean nobody is looking very hard.

Then you have testers, but you may not have much independent testing.

Some reporting is part of the job. If nobody knows what has been tested, what is failing or what the release risk looks like, the test manager is not helping by refusing to update anything.

That is a problem. It does not automatically mean the testers are doing nothing, but a function nobody can explain is going to lose trust eventually.

Maybe. But if the supplier being paid to provide testing is also the sole author of the report explaining how well the testing is going, a little healthy scepticism is not unreasonable.

Paying a consultancy a fortune does not buy defect-free software. If someone promised that, keep the slide deck; it may become useful evidence later.

Possibly. Length is not the crime. A complicated programme may genuinely need a detailed strategy. The problem is when nobody can explain what will be tested differently because the document exists.

It should look at the testing. This sounds obvious until you see supplier reviews that spend most of their time scoring governance templates and whether the weekly report was submitted on Tuesday.