An online store that a blind customer cannot buy from is no longer just a design problem. Since June 2025 it is a legal one, in Europe and in the United States both, and it tends to arrive as a demand letter rather than a bad review.
Most owners hear that and reach for the fastest thing on the market: a widget that drops into the page, adds a small accessibility button, and promises compliance in one line of JavaScript. The litigation data says that is the wrong purchase.
The European Accessibility Act (Directive (EU) 2019/882) came into force across the EU on 28 June 2025. Two details matter commercially. It treats e-commerce services broadly, covering not just the checkout button but every step that leads to a sale. And it applies to any provider selling to consumers in the EU whether or not that provider is based in the EU, so a US or UK store shipping into Germany is inside the scope. It also expects a public accessibility statement explaining how your service meets the requirements.
The American pressure is older and more expensive. UsableNet's midyear 2026 data puts federal and state courts on pace for roughly 6,000 website and mobile app accessibility lawsuits this year, up 20 to 25 percent on 2025, and around 80 percent of those filings involve e-commerce companies.
Retail is the easiest target in the room. A tester opens your store, tries to buy something with a screen reader, and either finishes or does not. There is nothing to argue about afterwards.
WebAIM scanned the home pages of the top one million sites in February 2026 for its annual accessibility report. 95.9% of them had detectable failures against WCAG 2. Shopping sites averaged 71.0 errors per page against 56.1 across the whole sample, which is 26.6% worse than average. By platform, Shopify home pages averaged 75.1 errors and Magento home pages 75.8.
Here is the part most owners miss. The failures are boring and repetitive. Six issue types account for 96% of everything WebAIM detected: low-contrast text, images with no alternative text, form inputs with no labels, empty links, empty buttons, and a missing document language. A third of all form inputs in that sample were not properly labelled, which is precisely the failure that stops a screen reader user halfway through checkout.
Commercially, that is good news. The list of things wrong with your store is short. It is just long-lived, because nobody owns it.
About 20 percent of the companies sued in the first half of 2026 already had an accessibility widget or overlay installed, according to UsableNet's analysis of midyear litigation data published in July 2026.
That number is the whole argument. Overlays sit on top of the page and let a visitor adjust contrast, text size, spacing and similar display settings. They do not rewrite the markup underneath. Most accessibility claims are brought by people who are blind or use screen readers, and a screen reader reads the code, not the visual layer. If a size selector announces nothing, if the promo-code field has no label, if a payment error message is invisible to assistive technology, a toolbar in the corner changes none of it.
A widget can be one small part of a programme. Sold as the programme, it buys a monthly invoice and a false sense of safety.
In the stores we work on, accessibility failures are rarely spread evenly across thousands of pages. They cluster in a handful of shared components: the header and mega-menu, the layered navigation filters, the product options block, the cart drawer, and the checkout form. Fix those five and you have fixed most of the catalogue at once, because every product page renders the same code.
The second cluster is third-party scripts. Review widgets, live chat, cookie banners, and payment iframes all inject markup you did not write, and they are a common source of unlabelled buttons and keyboard traps. They also update on their own schedule, which is why a store that passed an audit in March can fail in June without anyone touching the theme.
That is the real shape of the cost. The first pass is a project, usually a few weeks of front-end work on a Magento or Shopify storefront rather than a rebuild. Keeping it is ongoing maintenance, the same as security patching. Treat it as a one-time audit and you will buy the audit again next year.
It is also the same engineering discipline that shows up in conversion rate work. Unlabelled fields, invisible error states and controls that cannot be reached by keyboard hurt every shopper on a bad connection or a small screen, not only the ones using assistive technology.
Start with the buying journey rather than the site. Search for a product, filter it, choose a size, add it to the cart, apply a discount code, and complete checkout, using only a keyboard and then a screen reader. Whatever breaks in that sequence is what a plaintiff's tester will find and what a real customer already found.
Then fix the shared components rather than individual pages, publish an accessibility statement you can actually stand behind with a monitored contact address, and add an accessibility check to the release process so the next promotion does not undo the work. For a store selling across several EU markets the statement and the testing need to hold in every language you sell in, which is the kind of detail that turns a simple task into a multi-market maintenance problem.
Avoid a promise you cannot keep. An accessibility statement claiming full conformance on a site that has never been tested with a screen reader is worse than no statement at all, because it is a written admission in a document the other side will read first.
Accessibility became a compliance requirement for online stores in both of their biggest markets, and the enforcement is aimed squarely at retail. The technical work is narrower than the panic suggests, roughly six recurring defects concentrated in a few shared components. What does not work is buying a widget and calling it done, which is exactly what one in five defendants had already tried.
If you are looking at an accessibility deadline with a storefront nobody has audited, this is the kind of work we do at Encomage. It usually starts with testing the actual buying journey on your store, since that is where the legal exposure and the lost orders sit in the same place.
Let’s build something powerful together - with AI and strategy.
