Between 2026-07-17 and 2026-07-17, an automated WCAG check scanned 429 websites. 52.1% of 326 sites fell in F, the lowest grade band with at least 10 sites. The sample is the home pages of 429 websites from two research sets built from a target list for the German market: online shops, transport and travel, banking and insurance, telecoms, football clubs, automotive, public-sector bodies and other organisations. A group is the label on our target list, not a finding that any law applies to a site. Most common detected failure mapped to a WCAG success criterion: Insufficient colour contrast (color-contrast), 88.0% of 326 sites. Second: Links with no discernible text (link-name), 34.0% of 326 sites. These are detected failures of an automated check in aggregated data, not a verdict on any site.

Three numbers

  • 52.1% of 326 sites in the grade breakdown fell in F, the lowest grade band with at least 10 sites. Scans from 2026-07-17 to 2026-07-17.
  • 88.0% of 326 sites in the rule breakdown had a detected failure of the most common rule mapped to a WCAG success criterion: Insufficient colour contrast (color-contrast). Scans from 2026-07-17 to 2026-07-17.
  • 34.0% of 326 sites in the rule breakdown had a detected failure of the second most common rule mapped to a WCAG success criterion: Links with no discernible text (link-name). Scans from 2026-07-17 to 2026-07-17.

Method

The sample is 429 websites, scanned between 2026-07-17 and 2026-07-17. The scanner loaded each site in Chromium and ran an automated check (WCAG 2.1 Level A and AA, the WCAG 2.2 target size criterion 2.5.8, and axe-core best-practice checks) with axe-core 4.12.1. Sites that did not render are not assessed: they have no grade and no rule result, so the grade and rule tables have smaller denominators. Automated testing catches roughly 30 to 40 percent of WCAG success criteria, not 30 to 40 percent of a site's failures. We publish aggregates only. The minimum cell size is 10 sites, and no site is named.

Method in detail: How Inkluso tests

Detected failures by criterion

Detected failures by axe-core rule and WCAG success criterion, sorted by share of sites. Denominator: 326 sites. A site can appear in several rows.
CheckRule IDWCAG success criterionSitesShare of sites with a detected failure
Insufficient colour contrastcolor-contrast1.4.3 Contrast (Minimum)28788.0%
Links with no discernible textlink-name4.1.2 Name, Role, Value11134.0%
Images without alternative textimage-alt1.1.1 Non-text Content6219.0%
Buttons with no discernible textbutton-name4.1.2 Name, Role, Value4614.1%
ARIA roles missing required child rolesaria-required-children1.3.1 Info and Relationships298.9%
Incorrectly structured listslist1.3.1 Info and Relationships288.6%
Page disables zoommeta-viewport1.4.4 Resize Text288.6%
Nested interactive controlsnested-interactive4.1.2 Name, Role, Value278.3%
ARIA attribute not allowed on this rolearia-allowed-attr4.1.2 Name, Role, Value226.7%
Form fields without a labellabel4.1.2 Name, Role, Value206.1%
ARIA roles missing the required parent rolearia-required-parent1.3.1 Info and Relationships154.6%
List items without a parent listlistitem1.3.1 Info and Relationships134.0%
ARIA attribute with an invalid valuearia-valid-attr-value4.1.2 Name, Role, Value123.7%
ARIA commands without an accessible namearia-command-name4.1.2 Name, Role, Value113.4%
Hidden elements keep keyboard focusaria-hidden-focus4.1.2 Name, Role, Value113.4%
ARIA input fields without an accessible namearia-input-field-name4.1.2 Name, Role, Value113.4%
ARIA attribute prohibited for this elementaria-prohibited-attr4.1.2 Name, Role, Value113.4%

Grade bands

Sites by grade band, in grade order A to F. Denominator: 326 sites. The grade band is Inkluso's score out of 100 for the automated check: A from 90, B from 75, C from 60, D from 40, F below 40. A grade band is not a WCAG conformance level, and a high grade means the automated check detected few failures, not that a site is accessible.
Grade bandSitesShare of sites
A195.8%
B4012.3%
C3510.7%
D6219.0%
F17052.1%

Sites by country

Sites by country. Denominator: 429 sites. Country is the jurisdiction recorded for the scan, otherwise the language of the scan; it is not a finding about where a site's operator is based. Countries below the minimum cell size are left out.
CountrySitesShare of sites
DE429100.0%

Sets

Sites by source set. Denominator: 429 sites. A site in both sets is counted in each, so the rows can add up to more than the denominator.
SetSitesShare of sites
scan-database39090.9%
scan-database-rerun5011.7%

Sites by group

Sites by target-list group. Denominator: 429 sites. Each site counts in one group; the table shows the number of scanned sites and their share of the sample, no outcome figure for any group, and groups with fewer than 10 sites are not listed.
GroupSitesShare of sites
Online shops23655.0%
Transport and travel6314.7%
Banking and insurance6214.5%
Telecoms194.4%
Football clubs184.2%
Automotive122.8%
Public-sector bodies102.3%

Download the data

The full aggregate file, long format, one row per cell: accessibility-data-2026.csv

Columns: dimension, key, sites, share_pct, source_sets.

What this data does not show

  • It does not show a manual test. Every result comes from an automated check, which covers only part of WCAG.
  • It covers only the home page of each site, as the scanner loaded it. No other page of any site was checked.
  • It does not show whether any site meets a national act or any other legal requirement. A detected failure of an automated check is not a legal conclusion about a site.
  • It is not a representative sample of all websites. The sample is the sets listed above.
  • 20 cells with fewer than 10 sites are left out, so the tables do not add up to every site in every dimension.