Blog

WCAG 2.2 AA Checklist: All 55 Criteria You Can Test Yourself [Free Download]

TestParty
TestParty
October 17, 2026

Last updated: October 17, 2026

This WCAG 2.2 checklist puts all 55 Level A and AA success criteria on one page as checkboxes β€” a WCAG 2.2 AA checklist you actually run, not another explainer of what the standard means. Every row names the criterion, its level, and a one-sentence test you can perform yourself, organized under WCAG's four principles: Perceivable, Operable, Understandable, and Robust.

Key numbers: WCAG 2.2 numbers 87 success criteria, but 4.1.1 Parsing is obsolete, leaving 86 in force β€” 55 at Levels A and AA (W3C). In 2026, 95.9% of the top one million home pages carried detectable WCAG failures, averaging 56.1 per page (WebAIM Million 2026). Automated tools reliably decide about 9 of the 55 A/AA criteria and catch 60–70% of issue volume; the rest needs manual testing (TestParty data, 100+ brands). TestParty has remediated 35 million-plus issues as of August 2026, spanning all 55 criteria below.

How do you use this checklist?

For what each criterion means and why regulators require it, see our WCAG 2.2 AA pillar guide β€” this page is the test-yourself companion. Test one instance of every key template β€” home, collection, product, cart, checkout, account, search β€” rather than every URL.

Run a scanner first: tools reliably decide about 9 of the 55 criteria outright, catching 60–70% of issue volume in minutes. Then work every remaining checkbox by hand β€” keyboard, screen reader, 200% zoom β€” since the other 30% is judgment a machine can't make. TestParty runs this sequence daily via automated scans plus monthly expert audits, since theme updates reopen closed criteria. A spreadsheet version of these 55 rows is also available as a free download.

Perceivable: does every visitor perceive your content?

Perceivable criteria test whether information and interface elements can be perceived through sight, sound, or another sense. Twenty of the 55 A/AA criteria live here, numbered 1.x.

☐CriterionLevelPass test
☐1.1.1 Non-text ContentATurn off images: does each one still make sense from its alt text?
☐1.2.1 Audio/Video-only (Prerecorded)AFind an audio- or video-only file: is a transcript provided?
☐1.2.2 Captions (Prerecorded)APlay a video muted: are captions synced to the dialogue?
☐1.2.3 Audio Description/Alternative (Prerecorded)AClose your eyes during a video: does the audio alone convey it?
☐1.2.4 Captions (Live)AACheck a live stream: are real-time captions available?
☐1.2.5 Audio Description (Prerecorded)AACheck a prerecorded video: is a full audio-description track available?
☐1.3.1 Info and RelationshipsAStrip the CSS: do headings, lists, and tables keep their structure?
☐1.3.2 Meaningful SequenceARead the page with CSS off: does content order still make sense?
☐1.3.3 Sensory CharacteristicsARead instructions: do they avoid relying only on shape, color, or position?
☐1.3.4 OrientationAARotate a phone: does the page work in both portrait and landscape?
☐1.3.5 Identify Input PurposeAAInspect name, email, address fields: do they carry autocomplete attributes?
☐1.4.1 Use of ColorAView the page in grayscale: is all information still conveyed?
☐1.4.2 Audio ControlAFind autoplay audio over 3 seconds: can you pause or mute it?
☐1.4.3 Contrast (Minimum)AACheck text against its background: at least 4.5:1 (3:1 large text)?
☐1.4.4 Resize TextAAZoom to 200%: does text stay readable with nothing lost?
☐1.4.5 Images of TextAAFind text baked into an image: could it be real, styled text instead?
☐1.4.10 ReflowAANarrow to 320 CSS px: does content reflow with no two-way scrolling?
☐1.4.11 Non-text ContrastAACheck icons and input borders against their background: 3:1 contrast?
☐1.4.12 Text SpacingAAApply extra line, letter, word spacing: does content display cleanly?
☐1.4.13 Content on Hover or FocusAATrigger a tooltip: can you hover over it without it vanishing?

Operable: can every control actually be operated?

Operable criteria test whether every control can be operated β€” by keyboard, by pointer, or without triggering seizures or timing pressure. Twenty of the 55 A/AA criteria live here, including three new WCAG 2.2 additions.

☐CriterionLevelPass test
☐2.1.1 KeyboardATab through the page: can you reach and operate every control?
☐2.1.2 No Keyboard TrapATab into a modal or date picker: can you tab back out?
☐2.1.4 Character Key ShortcutsAFind a single-key shortcut: can it be turned off or remapped?
☐2.2.1 Timing AdjustableAFind a time limit (session, quiz): can it be extended or disabled?
☐2.2.2 Pause, Stop, HideAFind anything auto-updating past 5 seconds: can you pause it?
☐2.3.1 Three Flashes or Below ThresholdAWatch every animation: does anything flash 3+ times per second?
☐2.4.1 Bypass BlocksALoad a page: does a skip link jump past nav to main content?
☐2.4.2 Page TitledACheck the browser tab: does the page have a unique title?
☐2.4.3 Focus OrderATab through the page: does focus follow the reading order?
☐2.4.4 Link Purpose (In Context)ARead a link out of context: does its text say where it goes?
☐2.4.5 Multiple WaysAACheck the site: is there more than one way to find any page?
☐2.4.6 Headings and LabelsAAScan headings and labels: does each describe what follows?
☐2.4.7 Focus VisibleAATab through the page: is there always a visible focus outline?
☐2.4.11 Focus Not Obscured (Minimum)AATab past a sticky header: is focus still at least partly visible?
☐2.5.1 Pointer GesturesAFind a swipe or pinch gesture: is there a single-tap alternative?
☐2.5.2 Pointer CancellationAPress and drag off a button before releasing: does it not fire?
☐2.5.3 Label in NameACompare a button's visible text to its accessible name: match?
☐2.5.4 Motion ActuationAFind a shake- or tilt-triggered feature: does a standard control also work?
☐2.5.7 Dragging MovementsAAFind a drag interaction: does a single tap or click do the same?
☐2.5.8 Target Size (Minimum)AAMeasure small tap targets: 24 by 24 CSS pixels, or spaced apart?

Understandable: is content and behavior predictable?

Understandable criteria test whether content and interface behavior are predictable, and whether errors are recoverable. Thirteen of the 55 A/AA criteria live here, including three more new WCAG 2.2 additions.

☐CriterionLevelPass test
☐3.1.1 Language of PageAView the page source: is the html lang attribute set correctly?
☐3.1.2 Language of PartsAAFind a passage in another language: does it carry its own lang attribute?
☐3.2.1 On FocusATab onto a control: does focusing it alone avoid a context change?
☐3.2.2 On InputAChange a field or select value: does it avoid auto-submitting?
☐3.2.3 Consistent NavigationAACompare navigation across templates: same order everywhere?
☐3.2.4 Consistent IdentificationAACompare repeated icons (search, cart): labeled the same way every time?
☐3.2.6 Consistent HelpACompare help links or chat across pages: same relative order?
☐3.3.1 Error IdentificationASubmit a form mistake: is the field named in text, not just colored?
☐3.3.2 Labels or InstructionsACheck every input: a visible label, not just a placeholder?
☐3.3.3 Error SuggestionAATrigger a validation error: does the message suggest a fix?
☐3.3.4 Error Prevention (Legal, Financial, Data)AAOn a legal or financial action: can it be reviewed or reversed first?
☐3.3.7 Redundant EntryAMove through a multi-step process: is entered data auto-filled, not re-typed?
☐3.3.8 Accessible Authentication (Minimum)AATry logging in: is there an alternative to solving or transcribing a puzzle?

Robust: does it work with assistive technology?

Robust criteria test whether content works reliably across current and future assistive technology, mostly through correct markup. Only 2 of the 55 A/AA criteria remain here β€” 4.1.1 Parsing was declared obsolete when WCAG 2.2 published.

☐CriterionLevelPass test
☐4.1.2 Name, Role, ValueAInspect a custom control with a screen reader: does it announce its name, role, and state?
☐4.1.3 Status MessagesAATrigger a status update without moving focus: is it announced to screen reader users?

4.1.1 Parsing is skipped on purpose: modern browsers now resolve HTML parsing errors identically, so the W3C removed it as redundant. A scanner still flagging 4.1.1 is running an outdated rule set.

How do you check forms specifically?

Forms fail more often than any other component because a missing label or blocked autocomplete field is invisible on screen but load-bearing for screen readers. Seven criteria decide almost every form failure β€” run them on every field, then again on checkout.

☐CheckCriterionPass test
☐Every field has a real label1.3.1, 3.3.2Click each label: does focus jump to its field? A placeholder alone isn't a label.
☐Required fields are stated in text3.3.2Look at a required field: is "required" in text, not just an asterisk or color?
☐Errors are identified in text3.3.1Submit a mistake: does the message name the field, not just turn it red?
☐Errors suggest a fix3.3.3Read the error: does it say how to fix it, e.g., "use MM/DD/YYYY"?
☐Autocomplete is wired up1.3.5Inspect name, email, address, card fields: do they carry standard autocomplete attributes?
☐Login has no unassisted cognitive test3.3.8Try logging in: an alternative to typing a CAPTCHA or copying a code?
☐Errors are announced, not just shown4.1.3Submit an error with a screen reader running: is it announced, not just visual?

In TestParty's audits of Shopify Plus checkouts, shipping-and-payment fails more of these checks than any template: a promo-code field blocking paste breaks 3.3.8, and a "card declined" message that never reaches a live region breaks 4.1.3. See the Shopify WCAG 2.2 mapping for cart and checkout specifics.

What tools run this checklist for you?

WAVE, Lighthouse, and axe automate the machine-decidable slice β€” about 9 of 55 criteria outright β€” largely why automated tools catch 60–70% of issue volume. None finishes the checklist alone; see which WCAG criteria a scanner can and can't test.

WAVE overlays flags on the rendered page but checks one URL at a time and won't walk checkout for you. Lighthouse runs inside Chrome DevTools and CI, scoring 0–100 against a subset of rules β€” a regression gate, not a conformance test. axe-core is the open-source engine under many other scanners, and what most teams wire into CI for automated WCAG testing.

TestParty's thresholds are Lighthouse 90+, WAVE errors under 5, axe violations under 3 β€” a regression floor, not a conformance claim. A perfect score still leaves half this checklist untested. Full methodology: our WCAG audit guide.

Frequently Asked Questions

How many criteria are in the WCAG 2.2 AA checklist? 55 total β€” 31 at Level A and 24 at AA. The numbered list runs to 87, but 4.1.1 Parsing was declared obsolete, leaving 86 in force. WCAG 2.1 had only 50 criteria at A and AA.

Does passing every checkbox make my site conformant? Not on its own. A conformance claim requires testing every applicable criterion across full pages and complete processes, per the W3C's conformance requirements. This checklist catches real failures fast; a formal claim is a separate, documented statement.

What are the six new criteria in WCAG 2.2? 2.4.11 Focus Not Obscured, 2.5.7 Dragging Movements, 2.5.8 Target Size (Minimum), 3.2.6 Consistent Help, 3.3.7 Redundant Entry, and 3.3.8 Accessible Authentication. Four need a human walking an actual flow; only 2.5.8 is comfortably automated.

Why doesn't this checklist include 4.1.1 Parsing? Because it no longer exists. The W3C declared it obsolete when WCAG 2.2 published, since modern browsers now resolve HTML parsing ambiguities identically. The numbered list still runs to 87, but only 86 criteria β€” 55 at A and AA β€” are in force.

Built with TestParty's cyborg approach β€” AI-powered research combined with human accessibility expertise. This article contains TestParty's editorial analysis based on publicly available information. We're an accessibility vendor with opinions informed by working with 100+ brands, and we encourage readers to do their own due diligence when evaluating any solution.

Stay informed

Accessibility insights delivered
straight to your inbox.

Contact Us

Automate the software work for accessibility compliance, end-to-end.

Empowering businesses with seamless digital accessibility solutionsβ€”simple, inclusive, effective.

Book a Demo