BLOG · TOOLS COMPARISON

Free Accessibility Testing Tools:
What Each One Actually Catches

WAVE, axe DevTools, Lighthouse, Accessibility Insights and automated scanners compared — coverage, strengths, blind spots, and how to combine them into a workflow that holds up.

Updated August 2026 · Reading time: 8 minutes

Why More Than One Tool

Automated tools detect roughly 30–40% of WCAG failures — the same 30–40%, mostly. Running three overlapping scanners gives you false confidence without new coverage. The strategy that works is layering tools with different blind spots: DOM-level analysers for code defects, visual tools for contrast and focus, a real screen reader for the experience, and manual keyboard testing for interaction. Below: what each free tool genuinely contributes.

The Tools, Compared

🌊 WAVE (WebAIM)

Browser extension annotating the rendered page: errors, contrast, structural elements. Best for seeing problems IN CONTEXT on the page. Blind spots: no guided fixes, dynamic content needs re-triggering.

🪓 axe DevTools (Deque)

Rule engine used by most CI pipelines; precise selectors and WCAG references per violation. Best for developers fixing issues. Free tier covers core rules; advanced rules paid.

💡 Lighthouse (built into Chrome)

One-shot audit incl. accessibility score. Good baseline snapshot, but limited rule set and single-page only. Treat the score as a smoke test, never as conformance.

🔍 Accessibility Insights for Web

Microsoft's extension: FastPass (automated + tab stops) plus Assessment (guided manual evaluation). The best free structured manual-testing framework.

🗣️ NVDA / VoiceOver

Actual screen readers. Only way to experience what blind users get: heading navigation, form labels announced, focus order, alt-text quality. Automated tools cannot judge whether alt text MEANINGFUL, only whether present.

⚡ Automated URL scanners

Crawl multiple pages unattended and aggregate violations — right shape for site-wide baselines and regression checks. Ours scans 16 WCAG 2.1 AA rules per page and flags exactly where each failure lives.

What No Tool Catches

Four failure classes evade every automated scanner. Alt-text quality: a tool sees presence, not senselessness. Keyboard traps in custom widgets: only Tab-through reveals them. Focus visibility during interaction: static analysis misses CSS :focus styles removed at runtime. And comprehension: logical heading structure, meaningful link text, sensible error messages. Budget manual time for these — an hour of keyboard-and-screen-reader testing finds more real barriers than another three scanners.

⌨️ Keyboard pass

Tab through every interactive element. Can you reach, see, and operate everything? Escape closes modals? Focus never lost?

🗣️ Screen reader spot-check

Navigate by headings, fill the contact form, submit it. Anything confusing to you is worse for daily screen-reader users.

👀 Zoom test

200% zoom and 400% reflow: nothing clipped, no horizontal scrolling required to read.

A Workflow That Holds Up

A sustainable cadence for a small team: (1) automated URL scan across key templates monthly — home, product/detail, article, form, search; (2) axe DevTools during development on every new component; (3) keyboard pass before each release on changed flows; (4) full Accessibility Insights Assessment quarterly; (5) screen-reader session quarterly on checkout/signup. Document results each cycle — regulators and enterprise buyers increasingly ask for evidence of testing, and a dated log of scans and fixes is that evidence.

Going Deeper

Our EAA e-book includes a full tool-by-tool testing protocol mapped to each WCAG criterion.

Try Our Free Scanner →    EAA E-Book →

Frequently Asked Questions

What is the best free accessibility checker?

For a quick look: WAVE. For development work: axe DevTools. For structured manual evaluation: Accessibility Insights for Web. For whole-site coverage over time: an automated URL crawler. They answer different questions — most teams end up using two or three.

Can Lighthouse scores prove EAA compliance?

No. Lighthouse samples a subset of rules on one page load. Regulators look for EN 301 549/WCAG 2.1 AA conformance including manual criteria. A high score neither proves nor disproves it.

Do I still need manual testing if all tools pass?

Yes. Automated coverage tops out around 30–40% of WCAG success criteria. Alt-text quality, keyboard operability of custom widgets and comprehension require a human — see the section above on what no tool catches.

How often should we scan?

Monthly automated scans on core templates, plus a scan after every significant release. Quarterly manual passes. More frequent scanning costs little and catches regressions while the responsible code is still fresh.

Try Our Free Scanner →    EAA E-Book →