Stress-test a quality bar before the next sprint proves the standard is uneven, unmeasured, or quietly ignored by exception.
Quality bars fail when definitions are vague, when sampling hides the real defect rate, when exceptions recreate the old standard in private, and when throughput math ignores the review load you just added. A clean rubric slide is not evidence that the system can absorb a month of real edge cases without quiet shortcuts.
What to put on the table
One sentence for what "good" means, how you will measure it, who decides exceptions, and the rollback trigger if the bar stalls delivery or trust. Attach the draft rubric, sample defects, capacity model for review time, and the manager FAQ you will need within a week. If product, eng, and support disagree on the definition, stop and reconcile first.
Name the decision you will make if the stress test finds nothing new, and the delay criteria if tooling or ownership is not ready.
Attack surfaces
- Definition gaps: teams that can pass the bar by reading it differently.
- Measurement: samples that miss the failures customers actually feel.
- Throughput: review queues that recreate heroics and silent overtime.
- Exception load: managers inventing private passes that gut the standard.
- Incentive clash: goals that still reward speed over the new bar.
Optional customer seat if the bar changes what ships externally. Optional finance seat if quality spend or warranty cost is part of the story.
How to run it
Feed Pingpong the draft rubric, sample defects, and capacity notes. Early passes steelman the standard. Later passes attack from eng, support, product, and manager seats. End with a pass that turns surviving objections into a clearer rule, a named exception owner, or a hold. Delete invented "we will feel it" reviews and dual-counted quality savings.
When the bar changes release criteria or partner SLAs, force product and support seats to map every path that still assumes the old definition. Release checklists, status pages, and customer letters count. A bar that looks clean in a deck while launch checklists still use old gates will fail on the first contested ship.
Force a day-after narrative: what happens if a high-priority launch asks for a quiet exception, if two teams score the same defect differently, or if the review queue is already full. If those stories are stronger than your mitigation plan, fix the package before announce. Related: war-game a launch checklist, war-game a product principle, stress-test an SLA change, pretend you are the ops lead, and the war-game decisions hub. Process: how to run a Pingpong.