A flaky test costs more than another CI run. Someone has to decide whether to investigate, retry or merge. When that decision becomes guesswork, the suite stops being a useful release signal.
Start with your own evidence. For a few representative failures, record the original error, time spent investigating, whether the failure repeated and what actually fixed it. Keep infrastructure errors, test-data problems and failed product assertions separate. An industry-wide percentage cannot tell you which of those problems your team has.
The same applies when evaluating a new tool. Compare how quickly you understand a failure, not just how quickly the first test becomes green.
Zerocheck keeps the test revision, failing step, screenshots and attempts together. A pass after a supported transient retry stays labeled flaky; assertion failures are not retried. That gives the reviewer a result to inspect, while the team still owns the diagnosis.
Run your customer flows before merging, then inspect the failed step, screenshots and exact test revision.
Get a demo