LESSON VIEW
PQA-11 · LEVEL 3 AUTOMATION & RELEASE · PRACTICAL QA

The Go/No-Go: Release Decisions From Evidence

Objective: Turn test evidence into a release recommendation: exit criteria assessed honestly, open defects triaged by severity, residual risk stated openly — QA recommends with evidence, and the business owner decides.

⏱ ≈70 min full lesson · Practice pack: a release recommendation with three open defects · English edition — Español · العربية · বাংলা in translation

🧽 The practice business: FreshNest Home Cleaning — a fictional local cleaning company (owner: Maya Torres) invented for this course. Every example and exercise uses FreshNest, so you learn on a complete, realistic business without ever touching a real one.

⚡ Key Points

  • A release decision is an argument from evidence, not a feeling.
  • Entry criteria guard the start; exit criteria guard the claim 'we tested enough'.
  • Criteria are agreed BEFORE results arrive — afterwards, they bend.
  • Severity = harm × reach. Priority = when to fix. Different decisions.
  • Residual risk is stated openly: untested areas, accepted defects, unknowns.
  • QA recommends; the owner decides — the owner carries the risk. Exit ≠ zero defects.

📖 Read

A release decision is not a feeling. It is a short argument built from evidence: what was tested, what was found, what remains open, and what risk the business accepts if it ships. Criteria turn arguments into checklists. Entry criteria stop a team wasting final testing on a build that is not ready; exit criteria stop 'we tested a lot' from masquerading as 'we tested enough'. Criteria are agreed before the results arrive — moving the goalposts afterwards destroys their value.

Triage rates severity honestly with two questions: how much harm if a customer hits this, and how many customers will hit it? Severity is that combination — impact and reach, not how ugly a defect looks. Priority is a different decision: how soon to fix, weighing effort, workarounds and timing too. A defect can be high severity yet scheduled later because a workaround exists, or low severity yet fixed at once because it takes two minutes.

Residual risk is the honest remainder: the recommendation names what was not tested, which defects are knowingly shipping, and what happens if the risk materialises. Hiding it doesn't remove it — it removes the owner's chance to prepare. Who decides? QA recommends, with evidence; the business owner decides, because the owner carries the risk. And the misconception to drop: exit does not mean zero defects. It means zero unacceptable defects, and a written account of the rest.