ADA Title II, For state and local government
Fix your PDFs for screen readers
Upload your documents and get them back fixed, still PDFs, with a plain answer on what changed and what a person should check. From $0.45 a page. A free account includes your first 50 pages.
A whole folder at once: compare and download each file, and buy credits when you need more.
No account. One file at a time, and the file is not kept.
Example
What comes back, in colours
An example: a fictional Town of Sampleton agenda
- Still a problem: Some text is hidden from screen readers — see below
- Needs you: Title read from the largest type on the first page — check it
- Needs you: Language read from the document's text — check it
- Needs you: Marked the parts of each page — listen to check them
- Done: Title set
- Done: Shows its title, not the file name
- Done: Title also saved in the file's details
- Done: No reading-order jumps found
- Not needed: Words already readable
- Not needed: Form fields — none missing a name
✓ Green done automatically · ! Amber needs you or was not checked: look at it before you publish · ✕ Red still a problem · – Grey not needed
See the full answer for this example
Text a screen reader will skip
This text was marked as decoration, so a screen reader does not read it. That is right for a header or footer repeated on every page, and wrong for real content.
Which text was hidden
- 250 characters of this copy are marked as decoration, so a screen reader skips them, and they appear nowhere else on their own page (“ctional - MissionLayer demonstration)”, “own of Sampleton | City Council | R”, “ctional. Agenda - November 4 | P”) — on pages 1, 2. That is 14 percentage points more of the document than the original hid this way: text that used to be read aloud no longer is. If any of it is real content, do not publish this copy.
If it is real content: Do not publish this copy yet: the text would be lost to screen reader users. Fix it in a PDF editor first.
Still for a person to check
- The title “City Council Regular Meeting Agenda” was read from the largest type on the first page. Check that it says what the document is, and change it before you publish the copy if it does not.
- The parts of each page and a reading order were marked in this copy, where the original had none. Nothing automated can say whether each part is marked as the right kind — a heading as a heading, a list as a list — so listen to it with a screen reader, or ask somebody who uses one, before you publish it.
- 250 characters on pages 1 and 2 are marked as decoration, so a screen reader skips them, and they appear nowhere else on their page. Check that this text is only a header, footer or page number. If any of it is real content, do not publish this copy.
- Nothing here was checked by a person. Read the copy against the original before republishing it — especially any figure, name or date.
A fictional document, not a customer’s, and the answer GovAccess gave it, as it gave it. Your own files get their own answer, in the same colours.
Why your PDFs, and why now
The Department of Justice’s ADA Title II rule adopts WCAG 2.1 Level AA for the web content of state and local governments. That includes the PDFs your website offers, such as agendas, forms, notices and reports, with some exceptions for archived and older documents.
A scanned page has no words for a screen reader to read. A form whose boxes are named “Text1” tells a listener nothing, and a file with no title is announced by its file name.
- 26 April 2027
- Public entities with a total population of 50,000 or more, other than special district governments.
- 26 April 2028
- Public entities under 50,000, and special district governments.
Source: 28 CFR § 35.200(b), as amended by 91 FR 20902 (20 April 2026).
GovAccess does the repetitive part of the PDF work and tells you what a person still has to check.
From your PDF to a fixed copy, step by step
Create your free account
Your organization, your name and your work email. You start with 50 credits free, and 1 credit fixes one page.
See this step: Create your free accountAdd your PDFs
One file or a whole folder. GovAccess counts the pages and shows the credits needed before anything is fixed.
See this step: Add your PDFsPress Fix and watch
Each file comes back still a PDF, with a title, a language and names for form fields, and, where the whole file is a scan with no text, words under its pages where they can be read. Where the file, your account and the service allow it, also headings, lists and a reading order; where those are written and your account has picture descriptions on, a description of each picture, written by AI for you to check. A file that cannot be fixed costs nothing.
See this step: Press Fix and watchCheck what is amber
Compare each fixed copy with the original. Green (✓) was done for you; amber (!) is for a person to check before you publish.
See this step: Check what is amberDownload, and buy credits when you need more
Download one file or all of them. Credits come in packs, and you pay the invoice by card or bank transfer.
See this step: Download, and buy credits when you need more
Before you upload anything
Your files
The PDFs you fix, and the originals beside them, are kept a week after fixing so you can compare and download them, then removed. No other organization’s account can see them.
Where your documents goPictures and AI
Where a structure tree is written and picture descriptions are on for your account, AI may describe each picture, labelled as written by AI for you to check. Your organization can turn AI off altogether; what AI is sent is answered below.
Questions about AI and your dataPaying
Buy a credit pack and pay the invoice by card or bank transfer. A file that cannot be fixed costs nothing. Need a quote, a W-9 or a purchase order first?
Contact and procurementScanning a whole website, or several? Ask us.
What it is not
Not a widget on your website: nothing is installed and nothing is published for you. Not a certificate either: no automated tool can decide whether your organization meets the ADA, including this one.
What GovAccess does not do
Questions about AI and your data?
Can we say no to AI?
Yes. Anyone signed in to your organization's account can turn AI off for the whole organization on the Account page, or ask us and we will do it. You can also choose "Don't use AI on these files" when you send one set of files, and leave AI on for everything else.
With AI off, nothing from your documents or your website is sent to any AI company, and no AI model writes descriptions or other words of its own into your files.
Exactly what is sent to the AI, and when?
For a picture description: that one picture, the document's file name (for a web page version, the document's title instead: the one our layout reader read from it, or the one AI wrote for the version), the picture's caption if it has one, and any words already read inside the picture. Never the rest of the page. A picture found inside a scanned page is sent as that picture alone, with a note that it came from a scan, never the whole page. What comes back is a short description, and whether the picture looks like decoration. In a fixed PDF that is only a suggestion, and a person marks a picture as decoration in the review. In a web page version it is applied: a picture AI calls decoration is given no description there, so check the version before you publish it.
For a web page version AI writes: the document's text, page by page, and its file name. Its pictures go separately, as above, where picture descriptions are on.
For a worked example on a website finding: the public HTML of that one element of your website and the rule it failed. No document content.
Nothing is sent during a website scan, and nothing is sent unless somebody on your account asked for a fix, a conversion or a worked example.
Is our document used to train AI?
Under Anthropic's commercial terms, what we send through its API is not used to train its models. We do not use your documents to train anything either. How long Anthropic keeps what it receives is set by those same commercial terms. With AI turned off, Anthropic receives nothing from your documents at all.
Every question about AI, answeredHow we handle your documents
Free account, No card needed
Fix your first 50 PDF pages free
Upload a whole folder of PDFs and get each one back easier for a screen reader to read, with a plain list of what changed and what a person should still check.
After that, from $0.45 a page. Pay per page. No subscription. Credits don't expire.
- A whole folder at once, with the credits shown before anything starts.
- A file that cannot be fixed costs nothing.
- Credits never expire. Buy more by invoice, paid by card or bank transfer.
- Your originals are never changed. Each fixed copy sits beside its original to compare.
What happens next
- Fill in three boxes. No password.
- Add your PDFs and press Fix.
- Check what is amber, then download.
Want to see it work first? Fix one PDF now, no account