How it works — AdaReady

1. Templates, not pages

A catalogue of 10,000 products is not 10,000 accessibility problems. It is one product template with the same handful of issues repeated 10,000 times.

We discover your URLs from your sitemap, then cluster them by URL pattern and page structure. Testing runs against a sample of each cluster, and every finding records how many URLs its template represents. Your report says so plainly:

Scan #1182 28 Feb 2026

Discovered 10,412 URLs
Clustered into 38 templates
Scanned 124 pages across all templates
Always included: cart, checkout, search, login, 404

Checkout and account flows are scanned on every run regardless of clustering, with a real signed-in session. That is where the barriers that stop people buying actually live, and it is where automated tooling most often never reaches.

2. Two kinds of testing, labelled as such

Automated

Every template, every scan. Contrast, form labelling, landmark structure, name/role/value, target size, reflow at 320 px, forced-colours behaviour, motion preferences. Fast, repeatable, and good at catching regressions.

Manual

A person with a screen reader and a keyboard, working through your real flows. Whether alt text says anything useful, whether focus order matches the visual order, whether an error message tells you how to recover.

A person in headphones working at a laptop at a sunlit kitchen table, hands on the keyboard.

Automated testing detects roughly 30–40% of WCAG success criteria. Any vendor implying otherwise is selling you the easy 30%. Every finding in your report is tagged with the method that produced it, so you always know which is which.

3. The record is the deliverable

Reports are nice. What matters over time is the running history: every issue found, when it was found, what was done about it, who did it, and which scan confirmed it. Append-only, exportable, and yours.

It is also the thing that is worth something if anyone ever asks what you have been doing. Demonstrable, dated, continuous effort is a far better position than a badge in your footer.

A stack of thick deckle-edged paper on a black surface, marked with chartreuse index tabs, a dark green fountain pen alongside. — Reports are delivered as HTML and PDF, and every one is dated, scoped, and hashed so it can be cited later.

Discovered

Source, date, severity, WCAG success criterion, affected template and URL count.

Changed

What was altered, by whom, on what date, with a link to the change.

Verified

The scan that re-tested it, and its date. Or reopened, if it came back.

Also included

The visitor-facing pieces.

Preferences toolbar

A small script letting visitors set contrast, text size, spacing, and motion for themselves, remembered across pages. Roughly 4 KB on the pages nobody opens it on, because the panel only loads when used. It does not alter your conformance and we do not present it as though it does.

Statement and barrier reports

A published statement covering what was tested, when, by what method, and what is still open — generated from your live remediation record, so it cannot drift into fiction. Alongside it, a form for visitors to tell you when something blocks them, routed to you with a response clock.

A monitor on a pale desk in bright daylight, its screen showing an abstract arrangement of grey blocks with one chartreuse bar. — Findings, scan history, and your remediation record live in one dashboard. This is an illustration, not a product screenshot.

See it against your own site.

The free scan runs the same pipeline described on this page. You get the real report.

Request a free scan

Test the templates. Fix the source. Keep the receipts.