Every distinct page type, not every page. Ten templates covering four hundred pages is ten pieces of work.
KeevaathTech/Services/Accessibility conformance audit
LENS · Product assessment
Accessibility conformance audit
Tested the way somebody using assistive technology actually uses your product, and returned as a plan your team can schedule, not a list they will file.
What this is
An assessment of a live product or website against the Web Content Accessibility Guidelines, version 2.2, at Level AA. It returns a list of the barriers a disabled person actually meets, ranked by how badly each one blocks them, with the effort and cost of fixing each one attached.
It is not a scan. A scanner finds roughly a third of what matters and reports a great deal that does not. The remainder is found by operating the product the way somebody using assistive technology operates it, which means a person doing it.
Why this is being bought now
Three things happened at once. Public sector bodies in the United States came under a firm deadline through the Department of Justice rule on web and mobile accessibility under Title Two of the Americans with Disabilities Act. The European Accessibility Act brought private sector obligations into force across the European Union. And WCAG 2.2 added nine success criteria that most products built before 2023 have never been tested against.
Underneath the deadlines is the part that is easier to forget. Somewhere between one in five and one in four adults has a disability. A product that cannot be used by keyboard is a product a portion of your market cannot buy.
What gets examined
Templates
Journeys
The paths that matter. Registration, search, checkout, submission, account recovery. A barrier inside a journey is worse than the same barrier on a static page.
Forms
Labels, grouping, error identification, error suggestion, and whether a mistake can be corrected without starting again.
Documents
Any PDF or downloadable file that carries information available nowhere else. Frequently the worst offender and almost always excluded from other audits.
Interactive components
Menus, modals, carousels, date pickers, custom controls. Where most Level AA failures actually live.
Media
Captions, transcripts, audio description, and whether anything plays or moves without being asked.
How we test
- Automated pass first
Not because it is sufficient, but because it is cheap. It clears the obvious and tells us where to look harder.
- Keyboard only
Every journey completed without touching a mouse. Focus order, focus visibility, keyboard traps, skip mechanisms, and whether anything is reachable at all.
- Screen reader
At least one desktop and one mobile combination. What a page announces is often unrelated to what it displays.
- Zoom and reflow
Four hundred percent zoom, and a three hundred and twenty pixel viewport. This is where fixed layouts fall apart.
- Contrast and color independence
Measured, not judged. Including whether any information is carried by color alone.
- Motion and timing
Whether a reduced motion preference is honored, and whether any timeout can be extended.
What you receive
Findings, ranked by impact on a person
Each barrier described in plain language, with the success criterion it fails, where it occurs, what a user experiences, and a screenshot or recording where that helps. Ranked by whether it blocks a task, degrades a task or annoys somebody, because those are different problems and they do not deserve equal weight.
A remediation plan with effort attached
Grouped by root cause instead of by page, because one component fixed correctly usually closes twenty findings. Each group carries an effort estimate, so the plan can be scheduled against real capacity instead of admired.
A draft accessibility statement
Written for you to publish, describing the standard you are working to, what you have tested, what is not yet conformant, and how somebody reports a barrier. Both the European Accessibility Act and public sector obligations expect one.
A walkthrough
An hour with whoever will do the remediation. Reports get filed. Conversations get acted on.
Two things we do not do
We do not install an overlay. The widgets that promise conformance from a line of JavaScript do not deliver it. They have been the subject of a great many lawsuits, disability advocacy organizations have asked publicly that they not be used, and they frequently make a page harder to operate than it was. If you already have one, the audit will tell you what it is breaking.
We do not certify. Nobody can. There is no certification body for WCAG, and any firm offering a certificate is selling you a document with no standing. What we provide is an assessment of conformance at a point in time, against a named standard, with the method described so somebody else could repeat it.
What changes the price
Four things, in order of weight. The number of distinct templates. The number of primary journeys. Whether documents are in scope. And whether the codebase is available, because a finding we can trace to a component is worth more than a finding we can only describe.
A small marketing site with six templates and no documents sits at the bottom of the range. A multi-market application with twenty templates, eight journeys and a library of PDFs sits at the top. We quote a single fixed price once the questionnaire tells us which.
What we need from you
A staging or production environment we can reach, test credentials for any journey behind a login, a list of the templates and journeys you consider primary, and a named person who can answer questions during the work. If the codebase is in scope, read access to the repository.
Nothing else. We do not need a specification and we do not need your team to prepare anything.
Questions we are asked
How long does it take?
One to three weeks of elapsed time depending on scope, and we tell you which when we quote. The manual testing is the bulk of it and it does not compress usefully, because operating twenty templates by screen reader takes as long as it takes.
Is a scan not enough?
No. Automated tools reliably detect around a third of WCAG failures and produce a meaningful number of false positives. They cannot tell you whether a focus order makes sense, whether an announcement is useful, or whether a journey can be completed. Those require a person.
Do you fix what you find?
We can, under a separate engagement, and many clients prefer their own team to do the remediation with our plan in hand. The audit is deliberately sold on its own so that the findings are not shaped by who profits from the fixes.
What standard should we be working to?
WCAG 2.2 Level AA is the answer for almost everybody. It is what the Department of Justice rule references, what the European Accessibility Act effectively requires, and what procurement processes ask for. Level AAA is not expected and not always achievable.
We already have an accessibility widget installed. Does that help?
It does not, and the audit will show you where it is causing harm. Overlays sit on top of a page and attempt to correct it at run time. They cannot fix semantics, they frequently interfere with assistive technology the visitor already uses, and their presence has not prevented litigation.
Can we rely on the report in a procurement response or a contract?
Yes, within stated limits. The report names what was examined, what was not, the date, and who is entitled to rely on it. If a third party needs to rely on it, that is arranged in writing beforehand.
What if you find a great deal?
That is the usual outcome and it is not a failure. The plan groups findings by root cause precisely so that a long list becomes a short sequence of fixes. Most products need fewer than a dozen changes to close the majority of their findings.
Send the questionnaire and we will price it.
About fifteen minutes. A fixed price and a start date usually come back within two working days, from the person who would run the engagement, along with an honest read on whether we are the right team for it.