Perceivable. Operable. Understandable. Proven.
We test your website, app or documents against accessibility standards and tell your team what to fix.
Free 30-minute call. Free one-page audit. No commitment.
Digital accessibility testing and audit company, based in Delhi, India.
Standards we test against
What we test
Five kinds of thing, each worked over by hand on the platform it ships to.
Websites and web apps
The product as people meet it: real pages and real flows, in desktop and mobile browsers.
| Keyboard only | Every control reachable, in a sensible order, with a visible focus outline and no trap. |
|---|---|
| Screen readers | NVDA and JAWS on Windows, VoiceOver on Apple: names, roles, states, headings, tables and live updates. |
| Forms and errors | Labels, hints, error messages, and the order they are announced in. |
| Zoom and reflow | 200% text size and 400% zoom, down to a 320 pixel wide screen, with no sideways scrolling and nothing lost. |
| Contrast | Text, interface parts and focus outlines measured against the background they actually sit on. |
| Motion | Every animation answering the pause control, and the system's own reduced-motion setting. |
| Media | Captions, transcripts, audio description, and controls that work without a mouse. |
iOS and Android apps
Native builds, tested on real devices rather than simulators alone.
| VoiceOver and TalkBack | Focus order, custom gestures, navigation menus, and what each element announces. |
|---|---|
| Touch targets | Sizes and spacing, measured against the 24 by 24 pixel minimum and against platform guidance. |
| Text size | Dynamic Type on iOS, font and display sizes on Android, and magnification on both. |
| Orientation | Portrait and landscape, with nothing locked to one of them. |
| Motion and media | Reduced-motion behaviour, captions, audio description and media controls. |
| Contrast | Text and interface colours in the light and dark settings the platform offers. |
Components and design systems
Each part on its own, and again inside real pages, where context changes everything.
| Name, role, value | What the component announces, and the state it reports. |
|---|---|
| Keyboard and focus | Tab order, arrow keys, escape, where focus lands and where it returns. |
| States | Disabled, required, invalid, expanded, selected, busy and error states. |
| Target size | 24 by 24 pixels minimum, with spacing where the target cannot grow. |
| Tokens | Colour pairs, spacing and type sizes checked as tokens, so one fix covers every use. |
Design files
Before a line of code, so barriers never leave the drawing board.
| Colour pairs | Text, interface parts and focus indicators against each background they will sit on. |
|---|---|
| Type and spacing | Sizes, line lengths and text spacing at the largest setting the design allows. |
| Target sizes | Controls drawn big enough to hit, and spaced apart when they cannot grow. |
| Structure | Heading levels, reading order, and the landmarks the page will need. |
| Annotations | What each component must announce, and how it behaves on the keyboard, written for the developer who builds it. |
PDFs and documents
Documents people have to read, fill in or file, checked in the format you publish.
| Tags and reading order | The order a screen reader follows, and whether the text is real text. |
|---|---|
| Structure | Headings, lists, tables with header cells, and the language of the document. |
| Alternatives | Alt text for images, and descriptions for charts and diagrams. |
| Contrast and size | Text on its actual background, at the enlarged settings readers use. |
| Forms | Field labels, tab order, tooltips and error messages. |
| Metadata | Title, author and language, so the document is announced correctly before it opens. |
Every kind of product is tested against the same standards: WCAG 2.2 (Web Content Accessibility Guidelines), and where your market asks for it, ADA, Section 508, the European Accessibility Act, EN 301 549 and India’s RPwD Act. See all the standards we test against. We name the exact criteria and version in every report, and we never promise legal compliance.
What we do
Five stages, in the order we work: from the design file to the confirmed fix.
-
Stage 1
Design review
We read the design files before build: colour pairs, type sizes, target sizes, focus and hover states, reading order. Barriers cost least to remove here.
You get a list of the barriers in the file, and the token or component to change.
-
Stage 2
Test cases
We write the test cases for your product: every screen, every state, every input method, agreed with you before testing starts.
You get a test plan you can read, and a scope you can hold us to.
-
Stage 3
Audit and document testing
A full manual audit against WCAG 2.2, screen by screen, with real assistive technology. Documents and PDFs are tested for tags, reading order and structure.
You get every finding with its severity, criterion, steps to reproduce and fix.
-
Stage 4
Regression testing
Your team fixes, we go back over every issue in the report and confirm what is cleared. What is not cleared stays on the list.
You get written confirmation of each cleared issue, and what remains open.
-
Stage 5
Consulting and support
Guidance on fixes, compliance planning and training, so the next release starts from a better place than this one.
You get sessions with your team, and monitoring for as long as you ship.
Buyers asking for proof of conformance? VPAT and ACR reports come out of the same audit. Want a first look for free? The ADA compliance checker runs on one page of markup, in your browser.
Why choose Perceivio
Five reasons, each backed by how we work.
Manual testing comes first
Tools support our testers; they never replace them. Every issue in your report was confirmed by a person using assistive technology.
Certified testers, trained continuously
Our testers hold Trusted Tester certification. They train on new tools and AI-driven interface changes as they ship.
Reports developers can use
Each issue shows steps to reproduce, where it sits in code and how to fix it. Your team never has to guess.
Every fix gets retested
We check each fix after your team ships it and confirm it works with assistive technology.
Proof before you pay
The free one-page audit shows our method on your own product before you commit to a full audit.
How we test
People first, tools second. We name everything we use.
A missed barrier is the one that costs most, so we do not stop at one pass. Every product is worked over in several ways at once: automated checks, manual testing, real assistive technology, keyboard only, zoom and reflow, motion settings and media, on real devices and in real browsers. Where two methods disagree we keep looking, and the manual finding wins with the reason written down. An issue is only closed when we have tried to make it happen again and failed.
How one finding is proven
- Automated checks axe DevTools, Lighthouse and WAVE, for code errors.
- Manual pass A tester works through every screen by hand.
- Screen reader NVDA, JAWS, VoiceOver and TalkBack, reading the page in order.
- Keyboard only No mouse, in the order a keyboard reaches each control.
- Zoom and reflow Read at 320 pixels wide and at 200% text size.
- Voice control A spoken command reaches the control, and it answers.
One finding
- Criterion
- WCAG 2.2, 1.4.3 Contrast (Minimum)
- Severity
- Serious
- Written up
- Steps to reproduce, code location and fix
Cleared on retest only when we have tried to make it happen again and failed.
Specimen entry only, shown to explain how a finding is written up.
| axe DevTools | Automated checks in the browser |
|---|---|
| Lighthouse | Automated page checks |
| WAVE | Visual flags on the page |
| Accessibility Insights | Guided manual runs |
| Xcode Accessibility Inspector | iOS element checks |
| Android Accessibility Scanner | Android interface suggestions |
| NVDA | Windows screen reader |
|---|---|
| JAWS | Windows screen reader |
| VoiceOver | Apple screen reader |
| TalkBack | Android screen reader |
| Keyboard only | No mouse at all |
| Magnification and voice control | Enlarged display and spoken commands |
What you get
Reports your whole team can act on.
| Finding | Severity and criterion | Fix | Retest |
|---|---|---|---|
| Keyboard trap in menu | Critical2.1.1 | Trap-free focus order | |
| Low-contrast price text | Serious1.4.3 | Text pair at 4.5:1 | |
| Footer has no landmark | Moderate1.3.1 | Name the landmark |
Each issue lists severity, WCAG criterion, steps to reproduce, code location and fix.
How we rate each issue
Each finding in the specimen report carries one of these ratings. Select a rating to show only those findings, then select it again to show all of them.
- Blocks use for some people. Fix first.
- Breaks common tasks. Fix before release.
- Adds friction. Plan the fix.
- Small defect. Fix with nearby work.
Select a rating to show only the findings with that rating in the specimen above. Select it again to show all findings.
| Issue-by-issue audit report | Each issue lists severity, WCAG criterion, steps to reproduce, code location and fix. |
|---|---|
| VPAT and ACR on request | A completed conformance report your buyers and procurement teams accept. |
| Plain summary for managers | A short overview of risk and priorities for product and compliance leads. |
| Retest confirmation | Written proof that each fix works after we retest it. |
Free one-page audit
See our work on your own product before you pay.
We audit one page or screen by hand, send the findings and add a scope for testing the whole product. It shows what needs testing and with which methods.
- 1
Send one page
Share one URL or screen from your product.
- 2
We test it free
We audit it by hand and write up what we find.
- 3
You get findings plus scope
You receive the issues and a plan for the whole product.
Terms and conditions apply. Read the terms for the free offers.
How it works
From first call to confirmed fixes.
Read them in order. The line fills as each step is reached.
-
Book a free call
A 30-minute talk about your product and goals.
You get a written scope and a plan for what to test first.
-
Agree the scope
We list what to test and with which methods.
You get the methods named for each platform, agreed in writing before we start.
-
We test by hand
Manual testing with tools and assistive technology.
You get every finding confirmed by hand, with the steps to reproduce it.
-
Your team fixes
The report shows each issue and its fix.
You get a severity, a criterion and a fix for each issue, ready for your backlog.
-
We retest
We confirm every fix works.
You get written confirmation of every cleared issue, and what remains open.
Common questions
What is the difference between an audit and a VPAT?
An audit tests your product and lists every barrier with fixes. A VPAT (Voluntary Product Accessibility Template) is the standard form that reports the result, usually as an ACR (Accessibility Conformance Report). Most buyers need the audit first, then the VPAT.
What do you test?
Websites, web apps, iOS and Android apps, components and design systems, design files, PDFs and documents. If users touch it, we can test it.
Which standards do you cover?
WCAG 2.2, the ADA (Americans with Disabilities Act), Section 508, the EAA (European Accessibility Act), EN 301 549 and India’s RPwD Act. We name the exact version and criteria in every report. Dated changes sit on accessibility news.
Why is manual testing needed?
Tools catch code errors fast, but they cannot judge meaning, flow or real assistive technology behaviour. Our testers confirm every issue by hand with screen readers, keyboard only and magnification.
What do we receive at the end?
A report with every issue, its severity, the WCAG criterion, steps to reproduce and a fix. Plus a VPAT and ACR when you need one, and a retest of every fix.
How does the free one-page audit work?
Send one page or screen. We test it, send the findings and a scope for the whole product. No cost, no commitment.
Can we check our own pages before we book?
Yes. Our ADA compliance checker reads one page of markup in your browser and reports what failed, with the WCAG criterion beside every finding. It sends nothing and needs no account.
Book your free call
Tell us about your product in 30 minutes. We tell you what to test next.