One plan.
A selection tool that keeps selection out of the cheaper tier is not selling you anything, so there is no cheaper tier.
- Test selection on every branch and every pull request
- Flake detection, isolation reruns and the quarantine list
- The skip list, with the map entry behind each skip
- Unlimited repositories on one account
- Self hosted runners, with no separate tier
- Data retention for 400 days, exportable as JSON
The questions people actually ask
What counts as a developer?
Anyone who pushed a commit that Bellwether ran against in the billing month. Bots, release automation and anyone who only read the dashboard are not counted.
What happens when the map is wrong?
A test that should have run and did not is the failure mode that matters, so it is the one that is instrumented. Every skipped test is recorded with the map entry behind it, and a full run is scheduled nightly to catch drift. If a skip ever hides a real failure, the report tells you which entry was stale.
Does it need our source code?
No. The recorder runs inside your pipeline and sends file paths and test identifiers. Nothing leaves your runner except the map and the results.
Can we leave?
Delete the step. Your pipeline runs the whole suite again the way it did before, on the same day you decide, with no migration.
Anything not answered here, write to us. A person replies.