LESSON VIEW
PQA-12 · LEVEL 3 AUTOMATION & RELEASE · PRACTICAL QA

Full Cycle: Test FreshNest Like a Professional

Objective: Run the whole job once, in order — plan, scripted cases, an exploratory charter, defect reports, automation decisions, and the release call — and hand over an evidence pack a stranger could re-check without re-running your work.

⏱ ≈3 h capstone · Split across several sessions · Assessed against the published rubric · 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

  • The capstone is the whole job in order: plan → test → explore → report → automate → decide.
  • Plan first: requirements, money journeys, and an honest out-of-scope list.
  • Scripted proves the known; exploratory hunts the unknown. Budget for both.
  • Automate like an investor: stable × repetitive × rule-based, or don't.
  • The evidence pack is the deliverable: plan, matrix, notes, reports, decisions, recommendation.
  • A smaller, well-argued set with complete evidence beats a sprawling one with gaps.

📖 Read

The capstone is not a bigger test — it is the whole job, done once, in order: plan, test, explore, report, judge what deserves automation, and make the release call on the FreshNest practice site, producing the evidence pack a professional hands over at an engagement's end. Plan before you touch the site: requirements REQ-FN-01 to REQ-FN-06, the journeys that take bookings, the rules that involve money. Everything else is secondary; some things are consciously out of scope. A plan covering everything covers nothing well.

Mix scripted and exploratory work. Scripted cases prove known rules — prices, validation, wording. Exploratory sessions hunt the unknown — odd states, interruptions, real-device behaviour. The plan budgets for both; each finds defects the other misses. Then decide automation like an investor: automate checks that are stable (pricing rules change rarely), repetitive (every release needs them), and rule-based (the expected answer is arithmetic). Never automate one-off content scans, first-look visual checks, or anything still changing daily — the script costs more than the checking.

The misconception to drop: a capstone does not mean testing everything. It means justifying your choices — a smaller, well-argued test set with complete evidence beats a sprawling one with gaps you cannot explain. That standard is published in the rubric: coverage, evidence quality, defect reporting, judgement, honesty. You are assessed against it, not against other learners.