CONTROLLED SYNTHETIC FIXTURE
A finding you can inspect.
Reproduction steps, an observed result and the scope left untested. This is how a useful QA finding starts.
Discount lost after a payment retry
Review required. The expected final total was €80; the observed total was €100 on both execution attempts.
Finding CHECKOUT-001 · Owned local fixture · Executed 28 October 2026.
Suggested priority: High in this example. The displayed total loses the agreed discount after a successful payment retry. No real payment or customer impact occurred.
Reproduction steps
- Add the €100 headphones to the controlled store’s cart.
- Apply coupon
SAVE20; confirm the displayed total is €80. - Select the simulated declined test-card action; confirm “Payment declined.”
- Select the simulated valid-card retry action; confirm “Payment successful.”
- Check that the discounted total remains €80.
Observed: the final total was €100. Both execution attempts failed the final €80 assertion. The payment retry above is part of the journey; the automated test retry reran the entire check once. The fixture is deliberately broken; root cause was not autonomously established.

The saved report classifies this as suspected_product_defect. Repeated assertion failure is observed evidence; that label does not establish an automatically verified root cause.
Scenario results and untested scope
| Scenario | Recorded outcome | Decision |
|---|---|---|
| Healthy fixture | 4 passed, 0 failed | CHECKS PASSED |
| Deliberate checkout defect | 3 passed, 1 failed | REVIEW REQUIRED |
| Fixture unavailable before execution | 0 executed, 4 blocked | INCOMPLETE |
| Fixture identity changed after run | 4 passed, identity mismatch | INCOMPLETE |
| Fixture identity unavailable after run | 4 passed, identity unreachable | INCOMPLETE |
Every plan selected the same four checks: storefront availability, the initial €80 coupon total, discount survival after payment retry, and the quote API response. Account settings was excluded from the five-check catalog. The unavailable scenario started no browser or test attempts. In the deliberate-defect scenario, three passed and one failed. In the final two scenarios, all four selected checks passed, yet the overall decision remained INCOMPLETE because post-execution environment verification failed. Passing assertions alone cannot establish a complete trustworthy run when the fixture identity cannot be confirmed unchanged.
What the evidence establishes
Each failed attempt retained a screenshot, trace, browser log and error context. Passing checks retained appropriate logs and API observations, including the quote API response confirming €80. Raw artifacts contain private paths and are not published; only this synthetic screenshot and report summary are public.
- This bounded four-check example is not full application coverage or a complete four-week pilot report.
- The recorded browser was system Chrome 156.0.8078.48 with Playwright 1.63.0, using one desktop Chromium configuration. It was not a pinned browser image; video recording was disabled.
- No mobile matrix, cross-browser testing or accessibility certification was performed.
- This demonstrates local evidence collection on a synthetic store. It does not establish hosted customer integration, tenant isolation, production readiness, security certification or autonomous release approval.
- No real payment was processed. No customer outcome, business savings or guaranteed turnaround is claimed.
- A separate generated preview artifact exists but remains unapproved and unexecuted. It was not the source of these test results.
Agree your first QA scope
ESTECH offers a fixed-price one-time QA review (EUR 500 / US$570) and an optional four-week senior-led service from EUR 2,500 / US$2,850. Sentinel remains a controlled demo. Final scope, taxes and terms are agreed in writing.
Request a QA reviewOr email support@estechsentinel.com. Please send a product overview, not credentials or customer data.