The machine now runs your tests on every push; the last zoom out is deciding, on one page, what all of this effort is for. Twenty days of technique still needs direction, and direction is what a strategy is. Not a forty-page document nobody reads. Six answers.
A strategy is six answers, not a spreadsheet
- What are we protecting? The top five user flows that hurt the business when they break. Everything the suite does serves those. Nothing else gets an end-to-end test.
- Where does each check live? Unit tests for logic, written by developers, thousands of them, milliseconds each. Contract checks for the data agreements, like your Day 11 feed tests, fast and cheap. End-to-end tests for the money paths, few and sharp. Manual and exploratory work for judgment and visual truth. The layers exist because end-to-end is the slowest, most expensive place to learn anything; push every check down to the cheapest layer that can catch its bug.
- What runs when? Fast checks on every change a developer pushes. The full suite before anything merges into the main copy of the code. A nightly run for the slow flows. Feedback in minutes, not hours; you wired exactly this yesterday.
- How do we know the suite is healthy? The Day 17 numbers: pass rate, retry rate, duration, tracked over time. When retries climb, stop adding tests and fix the suite.
- Where does AI fit? Agents draft tests, tidy locators, and turn bug threads into tickets. They do not decide what deserves automation. That is a human call, and it is yours.
- What stays human? Judging that a screen is visually wrong, exploring for the risks nobody wrote a story for, and calling the release. Those do not get automated.
If you can write your team’s real answers to those six on one page, you are ahead of most teams shipping today.
Write yours for the store
Do it for Tumble Kitchen as if it were your product. You have lived every input: you know its money paths from Day 13, its layers from Day 11, its health numbers from Day 17, its CI rhythm from yesterday. Fifteen minutes, six headings, one page. It is the exercise that turns a test-writer into a QA lead, and tomorrow’s capstone is sharper with it in hand.
Two tools to take further. My Test Automation Strategy Builder turns answers about your real team into the one-page version, risk-ranked, with a starter test for your framework. And for the thinking behind the answers, I wrote a full guide to creating an effective test automation strategy, plus a companion on the people side, how to prepare your QA team for test automation, because a strategy nobody adopts is a document, not a strategy.
A strategy is six answers on one page. If it does not fit on a page, it is not decided yet.
Do this today
Write the six answers for Tumble Kitchen, on paper or in the strategy builder, and keep the page next to you tomorrow. The capstone is the suite your strategy says should exist. Tomorrow: the standards file, the final suite, the quiz, and your certificate.