01
Scan the domain
AAI starts by discovering pages and running a fast automated audit to identify obvious technical issues and recurring patterns.
A managed accessibility assurance workflow
How AAI Audit works
AAI is not just a scan. It combines automated checks, AI-assisted page analysis, manual testing, developer workflow, and re-checking so accessibility work becomes visible, manageable, and provable.
Prefer a quick visual overview?
Open the AAI Guided Walkthrough
See how AAI moves from organisation setup and domain verification through scanning, manual testing, developer actions, evidence, and reporting.
The workflow
From scan to testing, remediation, and re-checking.
The process is structured so your team can move from detected issues to confirmed actions, fixes, re-checks, and evidence.
Step by step
AAI turns accessibility into an organised workflow.
Each stage has a purpose. Automation finds fast signals. Manual testing validates what needs judgement. Developer actions make the work fixable. Re-checking helps prove progress.
02
Analyse the page context
AAI helps interpret page structure, forms, content patterns, and likely areas that need manual validation.
03
Validate with manual testing
Human testers check the areas automation cannot reliably prove, such as keyboard flow, instructions, form behaviour, and meaningful page experience.
04
Create developer actions
Confirmed issues are turned into practical developer work, grouped by page and issue pattern so teams are not buried in raw scan noise.
05
Re-test and re-scan
Manual failures are re-tested and automated failures are re-scanned, giving the organisation evidence that fixes worked.
06
Build an evidence trail
AAI preserves scan history, manual testing notes, developer actions, retest outcomes, and reporting context at page level.
What AAI gives you
A clearer path from risk to evidence.
The goal is not to overwhelm your team with findings. The goal is to organise accessibility work so it can be understood, assigned, fixed, re-checked, and reported.
- ✓Know which pages and patterns need attention.
- ✓Separate automated findings from human-validated issues.
- ✓Give developers clear actions instead of noisy raw results.
- ✓Re-test and re-scan fixes before claiming progress.
- ✓Maintain a page-level evidence history for reporting.