Skip to main content

Blog Online shops

Guide · E-commerce

Accessibility beyond the legal duty: who benefits, what the evidence shows and how to test your shop

Published on
Updated on
Reading time
20 min

Accessibility may matter to more of your customers than you think. We look at the figures, the studies and their limits, overlay tools, and a four-stage test of your own shop, ordered by effort.

You do not know which of your customers have poor eyesight, no free hand or trouble with small controls, and your shop will not tell you. It is safer to assume that some of them do.

The Bundesfachstelle Barrierefreiheit, a German federal accessibility body, quotes a rule of thumb from Aktion Mensch, a German organisation: accessibility on the internet is essential for 10 per cent of the population, necessary for at least 30 per cent and helpful for 100 per cent. It cites no study or method, so it is a rule of thumb, not a statistic.

Official statistics and surveys give an indication of the scale. In Germany, 9.4 per cent of the population had an officially recognised severe disability at the end of 2025. In the EU, 23.9 per cent of people aged 16 and over reported long-standing health-related limitations in 2024, and 37.0 per cent reported difficulty with at least one basic activity in 2022.

This guide sets out the numbers, what the research does and does not show, and a four-stage test of your own shop. The legal duties are covered in our merchant radar entry on accessibility in online shops.

How many people live with limitations?

The official statistic covers only the narrowest group. At the end of 2025, a little over 7.8 million people in Germany had an officially recognised severe disability, says the Federal Statistical Office (Destatis); people with limitations but no official recognition are not counted. Most severe disabilities develop only late in life, says Destatis, and a little over a third of this group (34 per cent) were 75 or older. For 4 per cent of them, blindness or a visual impairment was the most serious disability. A government bill of 1 April 2026 says only 3 per cent of disabilities are congenital. The bill is a draft, not law, and counts a broader group, so its figures cannot be combined with the 7.8 million.

Other surveys use broader definitions. Of people aged 16 and over in the EU, 6.7 per cent reported a severe limitation lasting at least six months; they are part of Eurostat's 2024 figure of 23.9 per cent. In 2022, 20.3 per cent of the same age group in the EU reported difficulty seeing. In the GEDA 2019/2020-EHIS survey by the Robert Koch Institute, 32.5 per cent of adults in Germany reported hearing difficulties and 5.4 per cent used a hearing aid. In the 2024/2025 Allensbach glasses study, 64 per cent of people aged 16 and over in Germany wore glasses, and 90 per cent of those aged 60 and over. Destatis puts the share of the population aged 65 and over at 23 per cent in 2025.

None of these figures can be added together or applied to your own customers. Eurostat asks whether people have difficulty seeing even when wearing glasses, so people who see well with them are not counted. The glasses study was commissioned by the Kuratorium Gutes Sehen (a German optical industry association), the German Association of Opticians and Optometrists (ZVA) and optical companies: bodies that sell glasses or represent those who do. One pattern is clear: in the EU in 2024, limitations affected 7.1 per cent of 16- to 24-year-olds and 72.3 per cent of people aged 85 and over. The W3C, the body that develops web standards, adds that some people with age-related limitations do not think of themselves as disabled, and that many older people have reduced contrast sensitivity, colour perception and near focus.

Who else struggles, if only for a while?

Not every limitation is permanent. The W3C describes temporary impairments, such as a broken arm or lost glasses. It also describes situational limitations, such as being in a place too loud to hear audio or in sunlight too bright to see a screen. People with a temporary impairment may not know how to use accessibility features.

Mobile shopping is full of such moments. The W3C says sufficient contrast helps people using a mobile device in bright sunlight or in a dark room, and that captions help in both noisy and quiet places. Its explanation of success criterion 2.5.8 of the Web Content Accessibility Guidelines (WCAG) 2.2 deals with the size of targets such as buttons and links, whether people tap or click them. It says targets that are large enough, or at least spaced far enough apart, help people using a device with one hand or on bumpy public transport. An older W3C document, last updated on 22 January 2013, notes that mobile users often turn the sound off in public places.

Our advice: take a phone outside, hold it in one hand, turn the sound off, then try to find a product and pay for it.

Recital 4 of the European Accessibility Act, Directive (EU) 2019/882, says that other persons with functional limitations, such as elderly persons, pregnant women or persons travelling with luggage, would also benefit. A recital explains but does not create an obligation. In its 2015 impact assessment, the European Commission wrote that large fonts, high-contrast displays and magnification make computers more comfortable to use for the large majority of the working-age population: a statement, not a measurement, and about computers, not websites.

What the Fribourg studies found

Earlier research on whether accessible design helps people without impairments was inconsistent: some studies found positive effects, others none. A 2017 doctoral thesis by Sven Schmutz at the University of Fribourg in Switzerland reports four studies conducted for the thesis.

In the first, published in Human Factors in 2016, 61 participants without impairments did tasks on versions of a municipal website in a lab. The version meeting WCAG 2.0 level AA led to shorter task times and higher completion rates than versions with low or very low conformance. But the participants were students, chosen because they were comparatively easy to recruit, and the weaker versions changed several features at once, from contrast to focus visibility. The study measured the combined effect of these changes, not the effect of any single feature.

A 2017 follow-up compared matched groups of 55 people with visual impairments and 55 without in a remote test. Higher conformance improved performance and ratings such as perceived usability in both groups, and the gain did not differ between them. In a third study, 110 younger and older adults without impairments used a laptop and a tablet in the lab. Compared with very low accessibility, high accessibility increased completion rate, speed and satisfaction. Age did not change the benefit.

The fourth study tested easy-to-read language with 128 participants on a leisure centre website. Comprehension improved, but participants liked the text less. The author advises caution where the main aim is to advertise a product or service.

What the studies do not show

The author is open about the limits. Only a government website and a leisure centre website were tested, and it is not yet clear whether the results apply to other kinds of website, including online shops. The studies took place mostly in the lab, and lab conditions can differ considerably from real life. Effect sizes fell from large in the first study, which involved students, to rather small in the third, which involved participants across a wide age range.

Our reading: in these tests, accessible design helped users without impairments. None of them tells you how large the effect would be in a real shop. Only your own shop can answer that.

Font, contrast and tap targets for older users

Several studies point to larger type and high contrast for older users, but many of them are far from recent. A W3C literature review on older users dates from 2008, is a working draft, and the W3C does not currently plan to update it. It summarises a 2001 study of 27 participants, mean age 70, which found font size the main factor in legibility on screen, and a 1998 study in which adults over 45 read better with high contrast and positive polarity (dark text on a light background).

A 2022 systematic review in Frontiers in Psychology, focused on mobile devices, concludes that older adults prefer larger fonts, but that readability declines beyond a critical size. It also found inconsistent methods across the studies it included. A 2012 conference paper from Fraunhofer Portugal found that older adults tapped less accurately on smartphone targets smaller than 14 mm. Its 40 participants were aged 65 to 95 and came from day centres in Porto, so the result does not apply to all older shoppers.

Who enlarges the text on their device?

Figures exist mainly for smartphones, and we found none for Germany. In August 2023, developers of the Chrome browser said around 40 per cent of Android users have a system font size other than the default. That includes people who chose smaller text, so the figure does not show how many made their text larger. The Dutch Appt Foundation reports that more than a fifth of the users it studied normally increase the text size on iOS and Android, and more than a third adjust it, up or down. Its data come from a software library built into various apps, not from a random sample, and cover the Netherlands only.

For desktop browsing, a W3C CSS Working Group explainer cites a 2018 finding that around 3 per cent of Internet Archive visitors had changed the browser's default font size of 16 pixels. None of these surveys captures people who would need larger text but never change a setting. The Wikimedia Foundation offers a hint: in a beta test of a new display menu, around 55 per cent of clicks by logged-in beta users were for a font size larger than the default, and the foundation then raised the default for logged-out readers.

Our advice, speaking as developers: make the default type legible for someone who never changes a setting. Build the layout so that it copes with larger system text and browser zoom without cutting off or covering prices, buttons or the basket. That is good practice, not a legal requirement.

Does accessibility raise conversion and revenue?

Our research turned up no study that measured a shop's conversion or revenue before and after accessibility work. There are surveys, and they need careful reading.

One of them is the UK Click-Away Pound survey. In 2016, 362 people took part, 280 of them from the UK, and 71 per cent of disabled customers with access needs said they would click away from a website they found difficult to use. In the 2019 report, 69 per cent of participants with access needs said they would simply leave a problematic website, and 86 per cent said they would spend more if websites were more accessible. Poor legibility was a problem for 56 per cent, up from 44 per cent in 2016, although a table in the report gives 55 per cent. Both surveys ran online; the authors say participants were entirely self-selecting and call the results estimates, not precise figures.

The headline figure of £17.1 billion for the UK in 2019 is not based on what respondents spent. It rests on an average of £3,460 per person: UK online retail spending divided by everyone over 16 with internet access. The Commission's 2015 impact assessment cites a Eurobarometer survey of 25,516 Europeans from March 2012. In it, two thirds said they would buy, or pay more for, goods and services that were more accessible and better designed for all. That is a stated intention, not measured buying behaviour.

So ask yourself: would you rather buy where the price is easy to find, the buttons are easy to hit and the checkout takes no effort, or in a shop you have to struggle through? You can trust your instinct on that. It is plausible that effortless use helps sales, but it is not proven. To find out for your own shop, compare basket and checkout abandonment before and after a change.

Do overlay tools help your customers?

European and German bodies alike take a critical view of overlays, add-on tools embedded in a website that offer accessibility functions of their own. In a joint statement of 17 May 2023, the European Disability Forum (EDF) and the International Association of Accessibility Professionals (IAAP) said overlays do not make a website accessible or compliant with European accessibility legislation, and can interfere with assistive technology and override user settings. The European Commission says overlays are not an appropriate solution if they do not ensure that the website itself meets the criteria of the European standard EN 301 549.

In Germany, the federal and state monitoring bodies for IT accessibility, which oversee public sector websites, said in March 2025 that adding an overlay does not necessarily make a website accessible and often makes it less accessible for people who use assistive technologies. The Bundesfachstelle Barrierefreiheit says in its guidance that overlays often create additional barriers, and that a shop cannot use an overlay tool to meet its digital accessibility obligations under the BFSG, a German accessibility law. BIK runs the BIK BITV test, a German conformance test for websites, and does not issue a statement of conformance for sites that use an overlay.

Your customers gain little: most overlay functions, EDF and IAAP say, copy features already built into browsers and operating systems. Our advice: respect the settings customers have chosen on their own devices rather than overriding them, and put the work into your shop's own templates, colours and controls.

What do automated tools miss?

One comparison of tools was carried out by the UK Government Digital Service (GDS). When GDS published what it found on its deliberately inaccessible test page on 24 February 2017, the page contained 143 failures in 19 categories. When only errors and warnings were counted, the best tool found 37 per cent of the barriers. When prompts for manual checks were counted as well, a different tool led with 41 per cent. All ten tools together picked up 71 per cent, and 42 barriers were missed by every tool. The GDS page with the results for each tool, last updated on 13 April 2018, lists 142 barriers and 13 tools: axe found 29 per cent (30 per cent with warnings) and WAVE 30 per cent (39 per cent with warnings).

Deque, the maker of axe, reported in 2021 that automated testing with its axe tools fully covered 57 per cent of accessibility issues on average, across more than 2,000 first-time audits whose data are not public. That is the vendor's own figure, and it cannot be compared with the GDS figures. GDS measured how many planted barriers were found; Deque measured the volume of issues in real audits.

The W3C, Google and WebAIM agree on the principle. The W3C says tools cannot determine accessibility, only help to do so, and sometimes give false or misleading results. Google's Learn Accessibility course calls automated testing a great first step, though not all checks can be automated. WebAIM, the maker of WAVE, says only a human can determine whether content is accessible. Lighthouse, an automated testing tool from Google, lists manual checks of its own, such as confirming by hand that custom controls can be reached with the keyboard and show a focus indicator. GDS warns that a tool finding no issues does not mean there are none.

How do you test your own shop in four stages?

The stages are ordered by effort, not by quality: the first two are for shop owners, the third for developers, the fourth for specialists. Work through them in order until you know how urgently your shop needs attention. The descriptions are based on what the tool makers and the organisations mentioned here say. For more tools, see the W3C list of evaluation tools, which included 119 on 12 November 2026. Being listed does not mean the W3C endorses a tool, and the W3C does not verify the entries.

Stage one: only your shop's address, nothing to install

  1. WAVE online by WebAIM: enter a page address on the WAVE website; there is nothing to install. WAVE shows your page with embedded icons and indicators, and red icons mark errors that need fixing. It works best for someone who knows about accessibility, but non-experts can benefit too.
  2. Google's PageSpeed Insights uses Lighthouse, an open-source automated tool, to analyse a page address for performance, accessibility, best practices and SEO. In each category, 90 or above counts as good, 50 to 89 needs improvement and below 50 is poor. The Lighthouse accessibility score is a weighted average of audits that are either passed or failed.

Two limits apply. If WAVE reports no errors, WebAIM says, that does not mean the page is accessible. And the online check may not run all of a page's JavaScript, so for scripted content WebAIM recommends its browser extensions. Our advice: go straight on to stage two to check your basket and checkout.

Stage two: your own shop in the browser

This takes some patience. Go through the whole buying journey: home page, category page, product page, basket and checkout. The first two checks need no tool.

  1. Use only the keyboard along the whole journey, and check that the focus indicator is always visible.
  2. Increase the system text size and the browser zoom on desktop and phone, and check that prices and buttons stay visible and usable.
  3. The WAVE browser extensions for Chrome and Firefox run entirely in the browser and send no information to the WAVE server, so they can also check password-protected pages.
  4. The Bundesfachstelle, in its guidance for online shops (in German), names two browser extensions: ARC Toolkit, an automatic checker, and HeadingsMap, for the heading structure.
  5. The Easy Checks on the BIK für Alle website (in German), from the same project as the BIK BITV test, are simple tests developed by the W3C that you carry out yourself in the browser. The Bundesfachstelle notes that quick tests give only a rough impression.
  6. Chrome's developer tools include a Lighthouse panel powered by axe and can simulate several vision deficiencies.
  7. The Firefox Accessibility Inspector checks for contrast issues, keyboard navigation issues and missing text labels, simulates colour vision deficiencies and contrast sensitivity loss, and can show the tabbing order.
  8. TPGi's Colour Contrast Analyser, a desktop program, helps you judge the legibility of text and the contrast of visual elements such as graphical controls and visual indicators. The BIK BITV test lists it as a contrast tool.

Stage three: tools for developers

This stage needs web development skills or knowledge of the BITV test steps.

  1. axe-core from Deque Systems is a testing engine that fits into existing test environments, so accessibility tests can run alongside functional tests. It has rules for WCAG 2.0, 2.1 and 2.2 at levels A, AA and AAA and is released under the Mozilla Public License 2.0.
  2. Pa11y runs automated accessibility tests on pages from the command line or Node.js and is licensed under LGPL-3.0.
  3. BIK's BITV self-assessment (in German) is a free, web-based tool for anyone who develops accessible websites and wants to follow the requirements of the BITV test. The Bundesfachstelle points online shops to the BITV test steps as guidance, because BITV 2.0, the German ordinance on accessible IT for public sector bodies, also refers to EN 301 549. For a comprehensive test, BIK für Alle recommends the self-assessment or a commissioned BIK BITV test, both of which come from its own project.

This is how we test our own website, t10.de. We use an automated test suite of our own, built on the freely available axe-core library. A single violation fails the test for that page, and each one is reported with the rule, its severity, a help link and the elements affected. All tests in the suite have to pass, with no violations reported, before we continue work on the site. Our acceptance checklist also calls for manual spot checks of focus indicators and keyboard operation. We run Lighthouse through PageSpeed Insights. A scheduled job regularly checks the URLs in our sitemap and creates fresh reports where they are missing or out of date, by default for mobile and desktop.

Stage four: a specialist test

GDS advises combining automated tests with manual checking, an accessibility audit and user testing. Since 1 March 2021, each BIK BITV conformance test has been carried out by a single tester, backed by expanded external quality assurance. BIK says the self-assessment is not equivalent to such an independent expert test, because judging barriers correctly takes expertise and regular testing experience.

When should you call in specialists? If stage one already reports many errors, or stage three is beyond you, leave the testing to specialists who know the tools. When a shop is built or rebuilt, these checks belong in the sign-off before launch, not after. Our page on how we plan and build online shops describes our approach.

Frequently asked questions

Does a good Lighthouse score mean my shop is accessible?
No. The Lighthouse accessibility score is a weighted average of audits that are either passed or failed, and manual audits do not count towards it. Lighthouse lists manual checks of its own, such as keyboard access and a visible focus indicator for custom controls. A high score therefore says nothing about the points that need checking by hand.
Does the BITV self-assessment replace a test by a test centre?
No. The BITV self-assessment from BIK is a free, web-based tool for anyone who develops accessible websites. BIK says it is not equivalent to an independent expert test by one of its test centres, because judging barriers correctly takes expertise and regular testing experience. In the BIK BITV test, a single tester does the work, backed by external quality assurance.
Should product texts be written in easy-to-read language?
In a Fribourg lab study, easy-to-read language improved comprehension but reduced how much participants liked the text. The author advises caution where the main aim is to advertise a product, and across the Fribourg studies only a government website and a leisure centre website were tested. Our advice is to try easy-to-read versions on a few individual pages first.
Do older customers benefit more than younger ones?
In a Fribourg study with younger and older adults without impairments, age did not change the benefit of high accessibility. Older people are far more likely to live with limitations, though: in the EU in 2024, 7.1 per cent of 16- to 24-year-olds reported health-related limitations, but 72.3 per cent of people aged 85 and over.
Can accessibility be tested before the shop goes live?
Yes. The WAVE browser extensions run entirely in the browser and send no information to the WAVE server, so they can check password-protected pages. axe-core fits into existing test environments, so accessibility tests can run alongside functional tests. Once the shop is live, a second pass is worth doing.

Want to know where your shop stands?

We check your shop with automated tools and also by hand, and tell you what needs doing technically. Whether your shop meets the legal requirements is a question for your legal adviser.

Sources

  1. Population with disability (Data extracted in July 2025) (opens in a new tab)

    Eurostat (Statistics Explained) accessed

  2. Brillenstudie 2024/2025. Ergebnisse der Allensbacher Untersuchung zum Sehbewusstsein der Deutschen (opens in a new tab)

    Zentralverband der Augenoptiker und Optometristen (ZVA) accessed

  3. Pressemitteilung: Zum Zahnarzt ja, zum Sehtest nein? (28. Juli 2025) (opens in a new tab)

    Kuratorium Gutes Sehen e.V. (KGS) accessed

  4. Diverse Abilities and Barriers (Updated 25 June 2024) (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  5. Accessibility, Usability, and Inclusion (Updated 6 May 2016) (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  6. Online-Shops und das Barrierefreiheitsstärkungsgesetz (opens in a new tab)

    Bundesfachstelle Barrierefreiheit accessed

  7. Introduction to Web Accessibility (Updated 3 February 2026) (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  8. Understanding SC 2.5.8 Target Size (Minimum) (WCAG 2.2) (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  9. Disability-Friendly Website Design: Consequences for Nonimpaired Users and Users with Impairments (2017) (opens in a new tab)

    Université de Fribourg (Folia-Repository), Dissertation Sven Schmutz accessed

  10. Web Accessibility for Older Users: A Literature Review (W3C Working Draft 14.05.2008) (opens in a new tab)

    W3C Web Accessibility Initiative (WAI), WAI-AGE-Projekt accessed

  11. PSA: Browser text zoom on Android will now work like it does on desktop (Aug 2, 2023) (opens in a new tab)

    Chromium Project, Mailingliste blink-dev (Google Groups) accessed

  12. Accessibility stats: Font size (Last updated on Jan 27, 2023) (opens in a new tab)

    Appt Foundation (appt.org) accessed

  13. Accessibility stats (opens in a new tab)

    Appt Foundation (appt.org) accessed

  14. Explainer: meta tag for text scaling behavior (opens in a new tab)

    W3C CSS Working Group (csswg-drafts, GitHub) accessed

  15. Reading/Web/Accessibility for reading/Updates (opens in a new tab)

    Wikimedia Foundation (MediaWiki.org) accessed

  16. The Click-Away Pound Survey 2016 Final Report (PDF) (opens in a new tab)

    Click-Away Surveys Ltd. (Rick Williams, Steve Brownlow) accessed

  17. Click-Away Pound Survey 2016 - Final Report Summary (opens in a new tab)

    Click-Away Surveys Ltd. accessed

  18. The Click-Away Pound Report 2019 (Februar 2020) (opens in a new tab)

    Freeney Williams Ltd. (Rick Williams, Steve Brownlow) accessed

  19. Joint statement on accessibility overlays (17.05.2023) (opens in a new tab)

    European Disability Forum (EDF) und International Association of Accessibility Professionals (IAAP) accessed

  20. Web Accessibility (Abschnitt Accessibility overlays, letzte Aktualisierung 23.06.2026) (opens in a new tab)

    Europäische Kommission, Shaping Europe's digital future accessed

  21. E-Commerce: Hintergrundwissen (Online-Shops) (opens in a new tab)

    Bundesfachstelle Barrierefreiheit accessed

  22. Beschreibung des Prüfverfahrens (Web) (opens in a new tab)

    BIK BITV-Test accessed

  23. Web Accessibility Evaluation Tools List (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  24. axe-core README (opens in a new tab)

    Deque Systems (axe-core, GitHub) accessed

  25. axe-core LICENSE (opens in a new tab)

    Deque Systems (axe-core, GitHub) accessed

  26. WAVE Browser Extensions (opens in a new tab)

    WebAIM (Utah State University) accessed

  27. Accessibility features reference (Chrome DevTools) (opens in a new tab)

    Google, Chrome for Developers accessed

  28. Lighthouse accessibility score (opens in a new tab)

    Google, Chrome for Developers accessed

  29. pa11y/pa11y README (opens in a new tab)

    Team Pa11y (GitHub) accessed

  30. Colour Contrast Analyser (CCAe) README (opens in a new tab)

    TPGi (ThePacielloGroup, GitHub) accessed

  31. Werkzeugliste (Web) (opens in a new tab)

    BIK BITV-Test accessed

  32. Accessibility Inspector (opens in a new tab)

    Mozilla (Firefox Source Docs) accessed

  33. BITV-Selbstbewertung (opens in a new tab)

    BIK BITV-Test accessed

  34. What we found when we tested tools on the world's least-accessible webpage (opens in a new tab)

    Government Digital Service (GOV.UK), Accessibility in government blog accessed

  35. Accessibility tools audit results - Overview (opens in a new tab)

    GDS accessibility team (alphagov, GitHub Pages) accessed

  36. Selecting Web Accessibility Evaluation Tools (opens in a new tab)

    W3C Web Accessibility Initiative (WAI) accessed

  37. WAVE Help (opens in a new tab)

    WebAIM (Utah State University) accessed

  38. Interactive controls are keyboard focusable (opens in a new tab)

    Google, Chrome for Developers accessed

  39. Häufige Fragen (opens in a new tab)

    BIK BITV-Test accessed

  40. Testen von digitaler Barrierefreiheit (opens in a new tab)

    Bundesfachstelle Barrierefreiheit accessed

  41. WAVE Web Accessibility Evaluation Tools (opens in a new tab)

    WebAIM (Utah State University) accessed

  42. About PageSpeed Insights (opens in a new tab)

    Google for Developers accessed

  43. Easy Checks: Barrierefreiheit einfach testen (opens in a new tab)

    BIK für Alle (c/o DIAS GmbH) accessed

Share on LinkedIn

Related articles

Guide · E-commerce

Retrofitting a Shopware 5 shop for the legal duties of 2026 and 2027

With no update promised by shopware AG, Shopware 5 shops have to adapt to the new rules of 2026 and 2027 themselves. This guide shows where each duty sits in the shop code, from the withdrawal button to the GARAN label, and how we approach each one.

Case study · E-commerce

One QR code on every appliance: from EU obligation to service portal

Since 2021, every EU energy label has had to point via QR code to the European product database, EPREL. For an international household-appliance manufacturer, we turned that into more than a compliance box-tick: a QR platform that routes every scan to the right destination depending on context, and a multilingual service portal covering over 600 products, both living off the same data foundation.

Search the website

Popular topics

All results