Original research
The state of Dutch webshop accessibility, 2026
We ran our own scan of 7,514 Dutch online shops — not just their homepages, but the category, product and cart pages a customer actually has to get through. Here is what an automated test finds, and what it cannot.
Data collected September 22, 2026 · axe-core 4.13.0 · WCAG 2.2 A/AA rules
What we found
99.1%
of shops had at least one barrier an automated test can find
68.1%
had a critical barrier — a button, form field or image with no accessible name
61.2%
of shops with a clean homepage still had a barrier deeper in the journey
7,514
Dutch webshops scanned
Percentages are of confirmed webshops. The headline figure carries a 95% confidence interval of 98.9–99.3%.
A clean homepage does not mean a usable shop
Most published research scans homepages. A homepage is not where people buy. We scanned up to five pages per shop — the homepage, a category page, a product page and the cart — and the pattern is consistent: the further into the buying journey you go, the more likely something blocks you. Of the shops whose homepage came back clean, 61% still had a barrier on a page further in.
| Page | Shops tested | With a barrier |
|---|---|---|
| Homepage | 7,514 | 97.9% |
| Category page | 6,698 | 97.3% |
| Product page | 6,020 | 98.2% |
| Cart | 5,102 | 93.9% |
Barriers by shop platform
Shops built on different platforms did not come out the same. Shops running WooCommerce were least likely to have a critical barrier; shops running Magento were most likely. The gaps are wider than the confidence intervals, so this is a real difference rather than noise.
| Platform | Shops tested | With a critical barrier |
|---|---|---|
| WooCommerce | 2,284 | 53.5% (51.4–55.5) |
| Shopify | 1,976 | 68.1% (66–70.1) |
| Other or custom-built | 134 | 68.7% (60.4–75.9) |
| Magento | 3,087 | 78.7% (77.2–80.1) |
This measures shops built on each platform, not the platforms themselves. A scanner sees the finished page — the theme, the plugins and the developer who assembled the shop account for much of what it finds, and this data cannot separate those from a platform's own defaults. Every platform here had shops with no critical barrier at all.
Publishing a statement does not make a shop accessible
Only 154 of the shops we scanned had an accessibility statement linked from their homepage. Those that did were barely better off: 64.9% had a critical barrier, against 68.1% of shops with no statement at all. A statement is a document about a service, not evidence that the service works.
| Group | Shops tested | With a critical barrier |
|---|---|---|
| Statement found | 154 | 64.9% |
| No statement found | 7,360 | 68.1% |
Three fixes would clear most of it
Critical barriers cluster in a few rules. If every shop in the sample fixed just three of them — image-alt, label, button-name — 63.8% would have no critical barrier left. None of the three is a redesign; each is a missing name on an element that already exists.
image-alt
Images with no text alternative (WCAG 1.1.1)
label
Form fields with no label (WCAG 4.1.2)
button-name
Buttons with no accessible name (WCAG 4.1.2)
What fails most often
The same handful of rules account for most of what an automated test finds. Low contrast alone appears on 93% of shops. Each rule is listed with the WCAG success criterion it maps to.
| Rule | WCAG criterion | Share of shops |
|---|---|---|
| color-contrast | 1.4.3 | 93% |
| link-name | 2.4.4, 4.1.2 | 68.6% |
| target-size | 2.5.8 | 38.4% |
| image-alt | 1.1.1 | 33.3% |
| frame-title | 4.1.2 | 24.8% |
| label | 4.1.2 | 24.2% |
| button-name | 4.1.2 | 23.9% |
| meta-viewport | 1.4.4 | 20.6% |
How we did it
- We started from a public list of 8,713 Dutch webshops, visited each one once, and kept whatever the site returned on a first visit — no cookie banners clicked, no logins, no purchases.
- Up to five pages per shop: the homepage, a category page, a product page, the cart, and the accessibility statement or checkout page where one was linked.
- Every page was tested with axe-core against WCAG 2.2 A and AA rules. Best-practice rules, which are not part of the standard, were excluded.
- 7,571 sites responded, and 7,514 of those were confirmed to be shops — either by their platform or by having a working product or cart page. Every percentage on this page is of those 7,514.
- 517 sites blocked our scanner and 578 were unreachable. They are counted here rather than quietly dropped, and they are excluded from every percentage.
What this does not show
- Automated testing finds only part of what WCAG 2.1 AA requires. Keyboard traps, screen-reader flow, focus order and whether a page actually makes sense all need a person to check them.
- A shop with no findings here is not therefore accessible — it means an automated test found nothing on the pages we looked at.
- We tested four or five pages, not whole shops. A barrier on a page we did not visit does not appear here.
- The starting list comes from a third party (eaa-monitor.nl) and reflects the shops it knew about in September 2026. It is not a register of every Dutch webshop.
Use the data
The full aggregates — every percentage, count and confidence interval on this page — as a JSON file. Free to re-use with attribution. No shop is named in it.
Download the dataset (JSON)If your shop is in this sample
We do not publish individual shop names or results, and we will not. If you think something here is wrong, or you want to know what we found on your shop, write to [email protected]
See where your own shop stands
Run the same scan on your site, free. You get the barriers we find, each mapped to the WCAG criterion it breaks — no account needed.
Scan my shop — free