How BarriScan works
Three steps, no dashboard to learn, no contract. Below them is the shorter and more useful list: the things we will not do.
1. We test your rendered pages in a real browser
- Your page is loaded in headless Chromium and given time to settle, so JavaScript storefronts are tested as your customers actually receive them. A scanner that reads static HTML would report a React shop as clean — a false all-clear is worse than no scan.
- We run axe-core, Deque's open-source engine and the industry standard, against WCAG 2.2 Level A and AA — the level European law works to, mapped through EN 301 549.
- A free scan fetches exactly one page: the URL you submit, nothing else. A paid scan crawls up to 150 pages at no more than one request per second, respecting your
robots.txt, with an honest user agent that links back to a page explaining who we are. - Findings that the engine can only flag as uncertain are labelled requires human confirmation and are never counted as violations. Padding the number would make the report useless to the developer who has to act on it.
2. You get findings in plain English, with the evidence attached
- Every finding is written in language a shop owner can act on: what it is, who it locks out, what it costs the business, and a sketch of the fix — not a rule identifier and a link to a specification.
- Every visual finding carries a screenshot of your own page with the offending element boxed. No screenshot and no element selector means we do not ship the finding.
- Findings are de-duplicated and prioritised across the site, so "this one theme file breaks 240 pages" reads as one job, not 240 tickets.
- Each finding names the WCAG success criterion it maps to, and the report prints the exact engine, browser and rule-set versions used, so any result can be reproduced or challenged.
- Screenshots and page evidence are kept for a maximum of 90 days and are visible only to the person who ordered the scan.
3. We fix them at a fixed price, and prove it with a re-scan
- Within 24 hours of a paid scan you get a fixed-price quote drawn from the findings — from $499 for a quick-fix pack to $0 for audit plus remediation. All bands and what they include.
- The work is done by our own developers, in your theme or repository. The fixes are yours and they stay after we leave.
- Every band ends with a verification re-scan and a written fix log: what was broken, what changed, what the re-scan says now.
- If the job runs longer than we estimated, we absorb it. The fixed price is the trust product.
Scan one page free
Give us a URL and a work email. You get a plain-English PDF: every barrier automated testing can detect on that page, each one photographed on your own site, with what it costs to put right.
One page, no account, no card. Full-site scan (up to 150 pages): $49 — see pricing.
What we never do
| We never | Why |
|---|---|
| Sell, resell, bundle or recommend an overlay | The US regulator required an overlay vendor to pay $1,000,000 over its claims, and 800+ businesses running overlays were sued anyway. A script in the visitor's browser cannot repair your markup. The full record. |
| Issue a compliance certificate, badge or seal | No such certification exists from us or any vendor. We report findings and fix them; whether you meet the EAA is a legal judgement for you and your advisers. |
| Scan a site without an attestation | Every scan requires someone to confirm they own or represent the site. We also honour a permanent do-not-scan list and any robots.txt rule aimed at our crawler. |
| Quote a fine you would supposedly pay | Penalty figures are national maximums, always attributed and always "up to". No EAA fine has verifiably been collected anywhere as far as we can establish, and we are not going to invent one. The honest enforcement picture. |
| Pretend a clean scan means a clean site | Automated testing reaches roughly 30–57% of the problem depending on how you count. We keep a fixture page of real, documented failures that our own scanner reports as clean, and we publish it. Here it is. |
What a scan can and cannot prove
Automated coverage notice: this report was produced by automated testing (axe-core engine). Automated tools find only part of all accessibility barriers — independent studies measure roughly 30–57% of issues, depending on the site. A clean automated result is not proof of accessibility, and this report does not certify compliance with the EAA, EN 301 549, or WCAG. Full coverage requires a human expert audit, which we offer as a separate service.
The long version, with the studies and worked examples: what automated testing actually finds.
Keep reading
- Pricing — every number we charge, in public.
- The EAA — scope, dates, enforcement.
- About our crawler — what BarriScanBot does on your server.