How we work
No black box. This is the exact sequence every audit follows, what we need from you at each stage, and what lands on your side at the end.
-
Scope call — before anything is billed
A 45-minute call to identify your page templates, your critical user journeys and the platforms in scope. We ask what deadline you are working to and why, because a June regulatory date and an enterprise customer questionnaire lead to different priorities. You receive a written fixed-price quote within one business day. Nothing is billed until you approve it.
-
Kick-off — day 0
We confirm environments (staging or production), any test credentials we need, and who your single point of contact is. If we need accounts to reach the parts of a journey behind a login, we agree that here rather than discovering it mid-test.
-
Automated pass — days 1 to 2
Every in-scope template is scanned to establish a baseline and catch the mechanically detectable failures: missing alternative text, unlabelled form controls, contrast violations, invalid ARIA, heading structure problems. This is the cheap third of the work, and we treat it as a starting point rather than a result.
-
Manual pass — days 3 to 6
The part automation cannot do. Keyboard-only traversal of every interactive component. Screen reader testing with NVDA and VoiceOver, checking that custom components expose correct name, role, value and state. Reflow at 400% zoom and 320 CSS pixels. Reduced-motion and timing behaviour. Full transaction journeys completed end to end under each condition.
-
Triage and rating — day 7
Findings are deduplicated across templates, rated blocker / serious / moderate / minor, and grouped by root cause. This grouping is where most of the value sits: forty separate findings often trace back to three shared components, and fixing those three components closes all forty.
-
Report — days 8 to 9
The written deliverable: one-page executive summary, full issue register with screenshots and suggested fixes, and a remediation plan sequenced by impact against effort. Delivered as a document plus a CSV or Jira-ready export so tickets can exist the same day.
-
Walkthrough — day 10
A 90-minute session with your engineers. We work through the priority list, demonstrate the failures live so the impact is concrete rather than abstract, and answer implementation questions. Recorded, and the recording is yours.
-
Follow-up — 30 days
Email access to the auditor who did the work, for clarification questions as your team implements. Not a support subscription — just the recognition that questions surface during the fixing, not during the reading.
-
Retest — when you are ready
Optional and quoted separately. We re-run the original baseline, verify each fix, and issue an updated conformance position. This is the document you need before publishing an accessibility statement or answering a procurement questionnaire.
What we need from you
- A stable environment — staging is fine and often better, provided it matches production.
- Test credentials for anything behind a login, plus a test payment path if checkout is in scope.
- One point of contact who can answer questions within a day.
- Notice of planned releases during the testing window, so findings do not go stale mid-audit.
Confidentiality
We sign your NDA, or ours if you prefer. Findings are never published, and we never name a client without written permission — which is why the examples on this site are anonymised and clearly labelled as illustrative. Test data is deleted 90 days after delivery; see our privacy policy for the detail.
If we get it wrong
If a finding in your report turns out to be incorrect, tell us and we will correct the report and reissue it at no charge. If the whole deliverable does not match what we agreed in the scope, our refund policy sets out exactly what you are entitled to.
Ready to start with a scope call?
It costs nothing and takes 45 minutes. You will leave it knowing what an audit would cover and what it would cost.