Method

This page describes what happens during a scan, in order. The scan runs two rules engines (programs that test a page's code against a list of accessibility rules). It checks their findings against the browser's accessibility tree (the information the browser passes to screen readers: each element's role, name and state). Then it runs our own behavioral checks and lists what automation cannot decide. The counts on this page come from the running scanner, and the accuracy figures come from its published benchmark.