How an assessment works
What AccessFold does when it assesses a page, what it records, and what stays with a person by design.
1. The page is opened in a real browser
Each page is loaded the way a visitor’s browser loads it, at a desktop width and at a narrow width. A page scan clicks, types and submits nothing; a journey you record replays your own steps and submits nothing unless you confirm a test environment.
2. Automated rules run
An open-source accessibility rules engine runs the rules tagged for your chosen WCAG version. Every rule reports a result: a problem, a pass, needs review, or not applicable on this page. Best-practice rules run separately and are advisory only.
3. AccessFold browser checks run
Four checks of our own look at keyboard traversal, open dialogs, the accessibility tree, and reflow and text spacing at narrow widths. They produce evidence for a reviewer; they do not decide criteria on their own.
4. Coverage is recorded
Every report states what was in scope — pages, baseline, viewports and checks — and what was not. A check that could not run is reported as not run, never as passed. A page we could not reach is reported as unreached.
5. Findings become issues
Findings are grouped by WCAG criterion and tracked as issues across runs, so a fix can be confirmed by re-assessment and a regression can be noticed.
6. Evidence is mapped, not re-judged
The same WCAG evidence can be cross-referenced to laws and standards that draw on WCAG. That mapping is a relationship; it does not create a second verdict.