A Guide to Testing and Benchmarking Shopify Store Accessibility

Understanding accessibility law is only the first step. The next, and more practical, challenge is determining how your website performs against recognised standards.

Many organisations mistakenly assume that installing an automated tool or adding an accessibility widget is sufficient. It is not.

Effective accessibility benchmarking requires structured testing, manual review, documentation, and an ongoing improvement plan.

At Radiant, we can deliver an accessibility review to highlight areas that may be limiting its accessibility, and actions to take in order to fix them. More detail on what our review covers is in the article below. If you’d like to inquire about having an accessibility review, get in touch here.

This guide explains how to:

  • Define your compliance target

  • Conduct automated and manual testing

  • Benchmark against WCAG

  • Prioritise remediation

  • Produce defensible documentation (UK and US)

  • What Radiant’s accessibility review covers

Step 1: Define Your Compliance Target

Before testing, clarify what standard you are measuring against. This article looks at the different accessibility compliance frameworks for the UK and US. Here’s a brief recap:

UK Organisations

  • Private sector → WCAG 2.1 AA strongly recommended under Equality Act

US Organisations

  • Private businesses → WCAG 2.1 AA best practice for ADA risk mitigation

Without a defined benchmark, testing results lack context.


Step 2: Run Automated Accessibility Testing

Automated testing tools analyse website code for detectable WCAG failures.

What Automated Tools Detect

  • Missing alternative text (alt text)

  • Colour contrast failures

  • Improper heading structure

  • Missing form labels

  • ARIA misuse

  • Empty links or buttons

Limitations

Automated tools typically detect only 30 to 40% of accessibility issues. They cannot reliably assess:

  • Meaningful alt text quality

  • Logical reading order

  • Keyboard usability

  • Cognitive clarity

  • Contextual instructions

Automated testing is a starting point, but not a compliance guarantee.


Step 3: Conduct Manual Testing

Manual testing is essential for meaningful benchmarking.

Keyboard-Only Testing

Navigate the site using only the Tab, Shift+Tab, Enter, and arrow keys.

Check for:

  • Logical focus order

  • Visible focus indicators

  • No keyboard traps

  • Accessible menus and modals

Keyboard failures are common sources of legal complaints.


Screen Reader Testing

Test with:

  • NVDA (Windows)

  • VoiceOver (macOS/iOS)

Evaluate:

  • Reading order

  • Landmark regions

  • Heading hierarchy

  • Form field labels

  • Error announcements

This reveals issues automated tools cannot detect.

Step 4: Visual and Design Testing

Colour Contrast

Minimum contrast ratios:

  • 4.5:1 for normal text

  • 3:1 for large text

Text Resizing

Content should remain functional at 200% zoom.

Responsive Accessibility

Test mobile accessibility separately (many lawsuits involve mobile barriers).


Step 5: Test Forms Thoroughly

Forms generate a high proportion of accessibility complaints.

Check:

  • Clear labels

  • Programmatically associated inputs

  • Descriptive error messages

  • Logical tab order

  • Accessible alternatives to CAPTCHA

If users cannot complete a transaction, risk increases significantly.

Step 6: Create a Benchmarking Matrix

Document findings using a structured framework and categorise issues:

  • Critical – Blocks access (e.g., unusable checkout)

  • Serious – Major barrier

  • Moderate – Usability friction

  • Minor – Cosmetic or low-impact

Prioritise the highest level failures first.


Step 7: Build a Remediation Plan

Phase 1: Quick Wins

  • Add missing alt text

  • Fix contrast errors

  • Correct heading hierarchy

  • Add form labels

These improvements often resolve a large percentage of detectable failures.

Phase 2: Structural Improvements

  • Refactor navigation

  • Rebuild inaccessible components

  • Improve keyboard operability

  • Redesign complex interactions

Phase 3: Governance & Monitoring

  • Quarterly automated scans

  • Annual manual audit

  • Accessibility included in procurement

  • Developer training

Accessibility must become part of ongoing digital governance.


Common Benchmarking Mistakes

  • Relying solely on automated scans

  • Testing only the homepage

  • Ignoring PDFs and downloads

  • Forgetting third-party plugins

  • Treating accessibility as a one-off project

Sustainable compliance requires continuous monitoring.

Accessibility as Risk Management

In the UK, documented audits demonstrate reasonable adjustment efforts. In the US, documentation can significantly reduce settlement exposure.

Accessibility testing is not just technical validation, it is legal risk mitigation and reputational protection.


Radiant’s Accessibility Review

We can conduct an accessibility review for your Shopify store - we’ll cover the following areas, and submit a report highlighting areas for improvement, and what steps you need to take to improve them:

Colour contrast & usage

  • Text vs background meets AA contrast ratios

  • UI elements (e.g. buttons, icons, form fields) have sufficient contrast

Typography & readability

  • Font sizes and line heights are legible across devices

  • Images are not used to display embedded text where it can be avoided

  • Error messaging is clear and communicated through text, colour, and visual styling

Interactions

  • Visible focus styles are in place for keyboard users

  • Hover, focus, and active states are clear and consistent

  • Touch target size is large enough for interaction.

Layout & hierarchy

  • Clear and logical heading hierarchy (H1–H6)

  • Content is grouped and structured in an intuitive way

Motion & animation

  • Review any use of excessive motion or animation

  • Reduced-motion preferences supported where feasible

Semantic HTML

  • Correct use of headings, lists, and landmarks (header, nav, main, footer)

  • Buttons are used for actions, links are used for navigation

  • Ensure id attributes are unique where used

  • Lists use correct HTML structure (ul, ol, li)

  • Tables use appropriate headers and structure where applicable

Page structure & metadata

  • Document has a <title> tag

  • <html> has a lang attribute with a valid value

Keyboard accessibility

  • All interactive elements are reachable via keyboard

  • Logical tab order throughout key templates

  • No keyboard traps

Forms

  • Inputs have properly associated labels

  • Required fields are clearly indicated

  • Validation and error messages are clear and accessible

Images & media

  • Meaningful alt text used for content images

  • Decorative images marked appropriately

Navigation

  • Navigation is consistent across templates

  • Skip-to-content link implemented where possible

ARIA (where required)

  • Used only where necessary and implemented correctly

  • No redundant or incorrect ARIA roles

Responsive behaviour

  • Content remains usable at different viewport sizes

  • No loss of content or functionality when zoomed to 200%

General checks

  • Content is clear and logical; areas of confusion are highlighted

  • Links and buttons are clearly labelled for their purpose

  • Review third-party apps for obvious accessibility barriers

What We Do Not Cover

  • Formal WCAG 2.1 AA certification

  • Full assistive technology testing across all devices

  • Legal interpretation of compliance risk

If you’d like to enquire about having an accessibility review for your store delivered by our team of experts, get in touch here.

Conclusion

Benchmarking website accessibility requires structured testing, realistic standards, and documented remediation.

Across both the UK and US, aligning with WCAG 2.1 AA, conducting regular audits, and maintaining improvement logs provides the strongest legal and operational position.

Accessibility is not about avoiding lawsuits. It is about building digital experiences that work for everyone, and proving that you have taken meaningful steps to do so.