Two categories of tool dominate the accessibility software market. Overlay widgets — accessiBe, UserWay, AudioEye — inject WAI-ARIA attributes at runtime and promise instant ADA compliance for a flat monthly fee, typically $490 to $1,490 per site per year. Real scanners — Pope Tech, Silktide, Monsido, axe-core, AccessProof — audit your actual rendered DOM and report issues you then have to fix. Real scanners cost ~10x less and ship the work back to your codebase, where it belongs. This guide is the version of the comparison we wish merchants had before signing one.
What overlays actually do
An overlay is a JavaScript snippet you paste into your site. It loads on every page, scans the DOM at runtime, and adds ARIA attributes (role, aria-label, tabindex, etc.) to elements it detects as missing them. Most overlays also expose a sidebar widget letting the user toggle visual settings — high-contrast, larger fonts, dyslexia-friendly typeface.
That part — the visual toolbar — is benign. It mimics what the operating system already provides and what every modern browser supports natively. The disabled-user research community has been clear for years that the toolbar duplicates capabilities most users either already have or do not want.
The other part — automatic ARIA injection — is where the harm is. Overlays cannot infer semantic intent from broken HTML. A <div onclick> styled as a button gets role="button" bolted on, but no keyboard handler, no focus state, no proper aria-pressed for toggles. A screen reader receives a button that is not really a button. A keyboard user receives a focus target that does nothing.
What real scanners do
A real scanner runs an audit engine — typically axe-core — against the actual rendered HTML. It enumerates violations by WCAG success criterion, severity (critical / serious / moderate / minor), and DOM node. The output is a list. You read it, you fix the source code, you re-scan.
Automated scanners catch roughly 30 to 50 percent of real WCAG issues — color contrast, missing alt, broken heading hierarchy, label-input pairs, ARIA misuse, target size. The remaining issues — keyboard traps, screen-reader logic, dynamic content, focus management — require manual review by someone with assistive-technology experience. Honest scanner vendors say this upfront. Overlays vendors claim 95 percent or more, sometimes "full compliance," through automation alone.
The six tools we benchmarked
The market splits cleanly into the two categories.
Overlay vendors
| Vendor | Mechanism | Pricing (small site) | Notable |
|---|---|---|---|
| accessiBe | JS overlay + AI-driven ARIA injection | ~$490–$1,490/yr | Largest vendor. Named in the Federal Trade Commission settlement (2024) over deceptive accessibility claims. |
| UserWay | JS overlay + user toolbar | ~$490–$890/yr | Owned by Audioeye since 2023. Same overlay-injection approach. |
| AudioEye | JS overlay + claimed human audit overlay | ~$590–$1,490/yr | Positions as "automation + human review." The human review is shallow on the smaller tiers. |
Real scanners
| Vendor | Mechanism | Pricing (small site) | Notable |
|---|---|---|---|
| Pope Tech | WAVE engine (real DOM audit) | ~$59–$199/yr (small) | Built by WebAIM partner. Honest scope: catches what automated rules catch. |
| Silktide | Proprietary scanner + page-level audits | ~$199–$990/yr | Stronger CMS-style reporting. Designed for content teams. |
| Monsido | Site-wide scan + accessibility + SEO + QA | ~$3,000+/yr | Enterprise-skewed. Bundle of compliance scanners. |
| AccessProof (us) | axe-core engine + monitoring + reports | $0–$49/mo | Free no-signup scan, monitoring, dated reports for defense file. |
What the courts have said
The legal picture has shifted decisively against overlays in the United States.
- FTC enforcement (January 2024) — the Federal Trade Commission concluded that accessiBe's claims of automatic ADA and WCAG compliance were deceptive. The company settled for $1 million and was prohibited from making similar claims.
- Active litigation — overlay-installed sites continue to be sued under ADA Title III. Courts have consistently rejected the argument that an overlay is a defense; the underlying HTML is what determines accessibility.
- The disabled community position — over 800 accessibility specialists have signed the Overlay Fact Sheet recommending against overlay use. Many screen-reader users actively block overlays.
An overlay does not protect you from an ADA lawsuit. If anything, it can complicate the defense: a plaintiff can argue the merchant knew about accessibility but chose a tool that does not actually fix the underlying issues.
Decision framework — how to pick
Three questions, answered honestly:
- Are you optimising for the appearance of compliance, or for the experience of disabled users? The first is short-term — until the next demand letter. The second is the only thing that holds up in front of a screen-reader user, a regulator, or a judge.
- Can your team or vendor fix the source HTML? If yes, you want a scanner. If absolutely not (legacy CMS you cannot touch, third-party platform with no edit access), the conversation is about platform migration, not overlay rescue.
- Do you need defense evidence? Dated reports of audits performed and remediation history are a real defense in an ADA case. A scanner produces them; an overlay does not.
If you would rather see what your site actually scores before deciding, our free WCAG scan runs the axe-core engine on any public URL, no signup, in under a minute. The same engine powers our paid plans. If the free scan returns 8 critical violations, an overlay does not make them go away — it papers over them.
Closing
The market has been mature enough for long enough that the comparison should not feel controversial. Real scanners cost less, do less marketing, and produce work that survives legal scrutiny. Overlays cost more, market more, and produce work that has now drawn FTC enforcement against the largest vendor. We bet on the scanner path. We think you should too.