Table of contents
Research context
A repaired button can pass its original test while the same component fails in another state or page. Retesting the reported example closes a narrow question. It is not evidence that the whole product conforms to an accessibility standard.
WCAG 2.2 provides testable success criteria. WAI guidance describes representative page selection and complete processes. It supports explicit coverage rather than a claim based on one clean screen.
Practical application
Method: preserve the original defect, criterion, steps, environment, assistive technology, and expected result. Retest the exact example, then sample other states and pages using the component. Record keyboard order, accessible name, state announcement, zoom or reflow result when relevant, and code revision.
Scope and inference limits: the evidence can support closure of the reported case and describe the sampled reuse. It cannot represent every disability, browser, assistive technology, content variation, or workflow.
Operating controls
Limitations: automated checks cover only part of WCAG, while manual results depend on tester method and environment. Keep untested combinations visible. The accessibility owner decides acceptance, broader audit scope, and release timing.
Sources
Frequently asked questions
What should happen first?
Begin with a bounded ticket, approved access, explicit acceptance evidence, and a named reviewer.
Who approves production changes?
The client's technical owner keeps final merge and release authority.
