Decide what not to test

When the test plan has grown beyond the time available.

Help me decide what to leave untested. I want a recommendation, not a balanced view.

For each item in my plan:
1. What would the failure cost, in money, trust or recovery time?
2. How likely is failure, given how the code is written and how often it changes?
3. Would a test here catch it, or is monitoring or a type system the better guard?
4. What does the test cost to write, run and maintain?

Then split into: **must test**, **test if time allows**, **do not test, accept the risk**, and **do not test, cover another way**.

Be willing to put things in the third group, and state the risk being accepted in plain words so it can be an explicit decision rather than an oversight.

Test plan and context:

Test strategy. Unit or integration, what to skip, fixture design, and pruning a suite that costs too much.