How a lab works
Each lab gives you a small, fictional site — Harbourline Physio, a three-location clinic that exists only in these pages — with a fault built into it on purpose. You read the evidence the way you would on a real engagement: rendered responses, logs, a test run, a health report. At each step you make a decision, and the lab tells you whether it matches the fixture's known answer and why. A wrong answer is explained, not hidden, and you can try again; your first answer is what the score records.
At the end you see the corrected version, the evidence that proves it, and where to take the same method on your own site: the guide the lab is drawn from, a generator that writes the checklist or prompt for your project, and — after all of that — the paid system that packages the workflow.
What "example verified" means
When your decisions match the lab's answers, the lab is recorded in My Projects as *verified against example*, with your first-attempt score. That wording is deliberate. It means you read a controlled fixture correctly. It does not mean your site has been checked, audited, or is ready for anything. Only running the checks on your own site establishes that, and the labs are careful to say so at every step.
The sample data
Everything in a lab is declared as sample data on the lab page, and the fixtures are public downloads so you can run your own tools against them. No real business, host, certificate or inbox is involved. Nothing you enter in a lab is transmitted; the record lives in your browser.
Fifteen minutes each
Four steps, four decisions, no account. Take them in any order; the Technical SEO and Conversion labs pair naturally with the migration and Playwright guides, and the Operations lab with the maintenance checklist.