WCAG 2.2: 2.2.6 Timeouts — The Complete Guide

Success Criterion 2.2.6 Timeouts is the AAA rule for the quietest way to lose work: doing nothing. If user inactivity can cause data loss, the user must be warned of the timeout's duration — unless the data is preserved for more than 20 hours even if they never lift a finger. In this deep-dive we watch an application form silently die of inactivity, then fix it with the official W3C sufficient techniques — warn of the duration up front, or preserve the data past 20 hours — and learn to test timeouts with keyboard, DevTools, and NVDA. What you'll learn: The exact requirement of 2.2.6: warn of the timeout's duration, unless data is safe for 20+ hours How 2.2.6 differs from 2.2.1 (control the timer) and 2.2.5 (survive re-auth) Live demos: a form that silently dies of inactivity — then the version that warns and confirms every save How to test: keyboard walkthrough of the warning dialog, DevTools inspection of the inactivity timer, NVDA procedure The official sufficient techniques: warn of the duration up front, preserve data past 20 hours, or set timeouts beyond 20 hours The 20-hour preservation exception — and why 'we keep drafts' must actually be true Official W3C techniques covered: 2.2.6 — warn of the inactivity duration at the start of the process 2.2.6 — the 20-hour rule: data preserved 20+ hours needs no warning 2.2.6 — set session timeouts beyond 20 hours of inactivity 2.2.6 — test inactivity, not just clicks Chapters: 0:00 — Title card: 2.2.6 Timeouts (Level AAA) 0:06 — Series intro 0:44 — What 2.2.6 means 2:21 — Broken vs fixed 3:54 — How to test: keyboard 4:55 — How to test: DevTools 6:08 — How to test: NVDA walkthrough 7:47 — How to fix: official techniques 9:33 — Checklist + next episode Links: Understanding SC 2.2.6 (W3C): https://www.w3.org/WAI/WCAG22/Understandin... How to Meet WCAG 2.2 — 2.2.6 quick reference: https://www.w3.org/WAI/WCAG22/quickref/#ti... NVDA screen reader (free): https://www.nvaccess.org/ Part of Understanding WCAG 2.2 — one deep-dive video for every success criterion, 1.1.1 through 4.1.3.