Deciding when to stop testing depends on the specific risks, budget, and quality standards of the system under test. Running all planned tests or fixing all faults is often unrealistic or insufficient, and running out of time is a constraint rather than a logical stopping criterion.