- Bodies measured
- 193
- Median score
- 17/100
- Measured
- August 20, 2026 - November 8, 2026
Accessibility of public bodies: the figures
What our automated checks found on the websites of the public bodies we measure, with the method and its limits. Free to quote.
Of 193 public bodies measured, 182 score below 70 out of 100. The median score is 17.
70 is our own reference point, not a legal threshold.
Totals
- Bodies measured
- 193
- Published with a score
- 192
- Score of 0, not named
- 1
- Median, all measured
- 17
- Median, without the zeros
- 17
Why some bodies are counted but not named
A score of 0 can come from a failed look at the site, such as a block page, rather than from the site itself. We cannot stand behind that number under a body's name, so these bodies are counted in the totals and in the first median, and are not named in this report. The second median leaves them out.
How the scores are spread
| Score | Bodies |
|---|---|
| 0 | 1 |
| 1-24 | 128 |
| 25-49 | 43 |
| 50-69 | 10 |
| 70-89 | 10 |
| 90-100 | 1 |
By sector
| Group | Measured | Median | Below 70 |
|---|---|---|---|
| government bodies | 73 | 23 | 70 |
| city governments | 58 | 12 | 54 |
| universities | 31 | 13 | 30 |
| cultural institutions | 11 | 16 | 11 |
| transport operators | 11 | 17 | 9 |
| health services | 9 | 25 | 8 |
By country
| Group | Measured | Median | Below 70 |
|---|---|---|---|
| Poland | 98 | 16 | 95 |
| United States | 56 | 21 | 49 |
| Germany | 39 | 15 | 38 |
A group with fewer than 5 bodies is not shown, so that no row points at a single body. Where one small group could be worked out from the others, the next smallest is left out too. Zeros are in these medians but are not counted per group.
Most common failures
Share of the 193 bodies with stored findings where the check failed at least once.
| Failure | Bodies | Share of bodies |
|---|---|---|
| 1.3.1 Info and Relationships | 181 | 94% |
| 4.1.2 Name, Role, Value | 128 | 66% |
| 2.4.4 Link Purpose (In Context) | 106 | 55% |
| 1.4.3 Contrast (Minimum) | 99 | 51% |
| 2.4.7 Focus Visible | 96 | 50% |
| 2.4.3 Focus Order | 94 | 49% |
| 2.5.8 Target Size (Minimum) | 82 | 42% |
| 2.4.1 Bypass Blocks | 74 | 38% |
| 4.1.1 Parsing | 66 | 34% |
| 1.1.1 Non-text Content | 63 | 33% |
Highest scores
Only bodies already public in the index are named here. The full list is on the index page.
| Rank | Body | Score | Measured on |
|---|---|---|---|
| 1 | USA.gov | 100/100 | November 1, 2026 |
| 2 | State of Illinois | 89/100 | November 3, 2026 |
| 3 | Portland | 87/100 | November 3, 2026 |
| 4 | Minneapolis | 79/100 | November 3, 2026 |
| 5 | Bremen | 78/100 | August 22, 2026 |
| 6 | Zarząd Transportu Miejskiego Warszawa | 77/100 | August 20, 2026 |
| 7 | Department of Education | 75/100 | September 17, 2026 |
| 8 | PKP SA | 72/100 | November 4, 2026 |
| 9 | University of California, Berkeley | 71/100 | November 3, 2026 |
| 10 | Kielce | 71/100 | August 26, 2026 |
Method and limits
- Standard
- WCAG 2.1 Level AA
- Scope
- One page per entity: the main public entry point a citizen lands on.
- Frequency
- Weekly re-measurement; the published index is the latest sweep.
- Engine
- Mixed: 183 of 193 entries were measured in a real browser; the rest were read at the source level, which cannot decide the criteria below.
Not decided by this measurement
- 1.4.3Colour contrast, against the colour actually painted behind the text
- 1.4.10Reflow: whether the page scrolls sideways at 320-375px
- 2.5.8Target size, which needs the rendered geometry
- 2.4.3Focus order, and whether a dialog keeps focus
- 2.4.7Whether focus is visible when it lands
- 4.1.2ARIA that JavaScript writes after the page loads
Neither passed nor failed here. Nobody is marked down for them, and a good score does not mean they were checked.
- These are automated checks only. No person tested these sites, and no assistive technology was used.
- A low score is not a legal finding. It does not say whether a body meets its legal duty.
- We chose the list of bodies. It is not a random sample, so the figures do not speak for all public bodies in a sector or country.
- Bodies were measured on different dates within the period shown, so the figures are not one snapshot of a single day.
- OAKMONT sells accessibility scanning and repair services. This index is our own work and has not been reviewed by an outside party.
How to cite
Oakmont Technologies LLC, "Accessibility of public bodies", automated measurements from August 20, 2026 to November 8, 2026, https://oakmonttechnologies.com/ranking/report
Data
Right of reply
Any body in the index can object, ask for a new measurement or ask to be removed, using the form on its own page in the index or by writing to us. A person reads every request.
Press contact
Questions about the figures or the method go to:
Oakmont Technologies LLC, 5511 Parkcrest Dr. Suite 103, Austin, TX 78731, United States
Measure it on your own site.
The free scan reads a page the way this register describes it, names what fails, and puts the criterion next to each finding.
No account. One page, about twenty seconds.