Article archive
Maybe. Bug count alone will not tell you. One catastrophic defect is more important than 150 spelling mistakes. Ten minor defects in a low-use admin screen may matter less than one intermittent failure in payment. Yet teams still put a giant ‘OPEN…
Enough to give you confidence that the change has not broken important existing behaviour. Not every test anybody has written since the dawn of the programme. Regression packs grow like loft clutter. Nobody wants to throw anything away because it might be…
Do not send a tester a link and say ‘test everything’. That sounds comprehensive and tells them almost nothing. A decent tester will work out a lot for themselves, but you will get more value if they know what the product is…
First: stop pretending you can recreate six months of missed testing in a heroic final week. You cannot. You can, however, stop wasting the time you still have. The worst response to late testing is usually a sudden obsession with paperwork. Test…
There is no magic percentage at which software becomes ready. Ninety-eight per cent of tests passing can be terrible. A much lower completion figure can be perfectly sensible if the important risks have been covered and the rest was deliberately deprioritised. The…
The main problem with testing your own software is not that you are a developer. It is that you know where all the bodies are buried. You know which field is mandatory. You know the order the journey is meant to run…
There is no useful universal price for testing a small app. Anyone who quotes one before understanding the product is either guessing or has a rate card they are very keen to use. A brochure site, a booking system and an app…
No. Not every piece of software needs a dedicated tester standing beside it before launch. Anyone telling you otherwise is selling testers rather than advice. A tiny internal tool used by five people, built by someone who understands it well and can…
Yes, it is possible. Testing is not free and more testing is not automatically better. Some organisations keep testing because nobody is willing to make the decision to stop. Another regression cycle feels safer than saying ‘we know enough’. A month later…
Possibly. A full test schedule does not mean useful testing is happening. Imagine a checkout service with 200 tests. One hundred and eighty cover field validation, page layout and minor combinations. Twenty touch payment. Then the system double-charges Visa customers. The dashboard…