Disclaimer: This guide is general information only - not legal advice. Accessibility obligations, thresholds and statutory-damages figures change. Before relying on this for compliance, confirm your specific duty and the current figures with a lawyer and against the official text of IS 5568 and the Equal Rights regulations.
In short: Israeli law requires documents you serve to the public to be accessible to WCAG 2.0 level AA, via standard IS 5568 Part 2 (Part 1 = websites, Part 2 = digital documents). An accessible PDF is not the same as a PDF/A: accessibility means a tag tree, logical reading order, alt text, a document language and title - structure a screen reader can follow. An inaccessible document can expose you to statutory damages of up to about NIS 50,000 without proof of damage.
Who this guide is for
- Israeli businesses and organizations that publish PDFs to the public - forms, price lists, reports, invoices
- Foreign companies operating in Israel that need to know what Israeli law requires from their documents
- Accessibility coordinators and web teams told to "make the PDFs compliant" and unsure what that means technically
- Designers and developers building documents that must pass an accessibility audit
What Israeli law actually requires
Israel's accessibility duty comes from the Equal Rights of Persons with Disabilities Law and its regulations (the service-accessibility regulations, 2013). The technical bar is set by the Israeli standard IS 5568, which is built directly on the W3C's WCAG 2.0 guidelines at levels A and AA - and AA is the legal minimum.
Two points that trip people up:
- It's WCAG 2.0, not 2.1. Many accessibility vendors' pages state "WCAG 2.1" - but the standard adopted into Israeli law is 2.0 level AA. The W3C's own policy page for Israel confirms WCAG 2.0 AA.
- The standard has two parts. Part 1 covers websites and web content; Part 2 covers digital documents - which is where PDF lives. So an accessible website with inaccessible downloadable PDFs has only done half the job.
"But my file is already PDF/A" - why that isn't enough
This is the single most common misunderstanding. PDF/A and accessibility are different things.
| PDF/A | Accessible PDF | |
|---|---|---|
| Purpose | Long-term preservation - the file looks identical in 20 years | A screen reader can read and navigate it |
| Guarantees | Fonts embedded, no external dependencies, fixed appearance | Tags, reading order, alt text, language, title |
| Standard | ISO 19005 | IS 5568 Part 2 / WCAG 2.0 AA / PDF/UA (ISO 14289) |
A file can be perfectly valid PDF/A and still completely inaccessible - for example, a scanned image saved as PDF/A is archival but unreadable to a screen reader. The two only meet in the PDF/A-2a / PDF/A-3a "accessible" conformance levels, which require tagging too. So converting to PDF/A is a good preservation step, but it does not, by itself, make a document accessible.
What makes a PDF accessible - the checklist
- ✅ A real text layer - the file contains text you can select and copy, not just a picture of the page. This is the most basic condition: with no text layer, a screen reader has nothing to read aloud.
- ✅ A tag tree (structure) - headings tagged as headings, paragraphs as paragraphs, lists as lists, tables as tables. This is what a screen reader navigates.
- ✅ Logical reading order - the tags follow the order a person reads, regardless of where things sit on the page.
- ✅ Alternative text on every meaningful image; decorative images marked as artifacts (skipped).
- ✅ Document language set (e.g. Hebrew) so words are pronounced correctly.
- ✅ A document title shown in the viewer (not just the file name).
- ✅ Tables with header cells properly marked, so a screen reader can announce "row 3, column Price".
The Hebrew / right-to-left catch
Hebrew (and Arabic) documents have an accessibility trap that Latin documents don't: the visual order of characters on the page and the logical reading order a screen reader needs are not the same. A PDF can look perfectly correct yet store its Hebrew in visual (reversed) order, so a screen reader reads it backwards or scrambled. Getting the reading order in the tag tree right - not just the appearance - is the part most generic "make accessible" tools get wrong for Hebrew. (If you've seen Hebrew come out reversed when copied from a PDF, that's the same underlying issue - see our guide on reversed Hebrew in PDFs.)
The legal risk
The accessibility duty is anchored in Regulation 35 of the Equal Rights of Persons with Disabilities (Service Accessibility Adjustments) Regulations, 2013. A person with a disability who encounters an inaccessible service can sue for statutory damages of up to about NIS 50,000 without proving actual damage (for websites, subject to a prior-notice requirement). For an organization that publishes many documents, that is a real, repeatable exposure - which is exactly why "make the PDFs accessible" lands on so many to-do lists.
How to make an accessible PDF - step by step
- Start from real text. If the file is a scan, run OCR first so it has a text layer. See making a scanned PDF searchable.
- Add a tag tree - mark up headings, paragraphs, lists and tables.
- Fix the reading order - especially for multi-column and Hebrew/RTL content.
- Add alt text to meaningful images; mark decorative ones as artifacts.
- Set the document language and a real title.
- Validate against PDF/UA (ISO 14289) with a tool like veraPDF, and test with a screen reader.
Tip: The most reliable way to get an accessible document is to start from an accessible source - a well-structured HTML page or template exported with tagging on - rather than trying to retrofit tags onto a flat PDF after the fact. Our Hebrew document templates are produced this way and pass PDF/UA-1 validation.
Common mistakes
- Assuming PDF/A = accessible. It doesn't (see above).
- Accessible website, inaccessible PDFs. Part 2 of the standard is about the documents, not just the site.
- Citing WCAG 2.1. The Israeli legal reference is 2.0 AA.
- A "scanned" PDF with no OCR. A picture of text is the least accessible file possible.
- Right-to-left reading order ignored. Hebrew that looks fine on screen can still read backwards to a screen reader.
More guides that help
Want to convert to PDF/A now?
With full Hebrew support
Frequently Asked Questions
Which WCAG version does Israeli law require - 2.0 or 2.1?
WCAG 2.0 level AA. The Israeli standard IS 5568 is built on WCAG 2.0 (levels A and AA), and AA is the legal minimum. Many vendor pages wrongly state 2.1 - the binding reference in Israel is 2.0 AA. Part 1 of the standard covers websites; Part 2 covers digital documents, including PDF.
Is a PDF/A file automatically accessible?
No. PDF/A is an archival format that guarantees the file looks the same in the future (fonts embedded) - it is about preservation, not accessibility. Accessibility is about structure a screen reader can follow: tags, reading order, alt text, a document language and title. A file can be valid PDF/A and still be completely inaccessible. The two overlap only in the PDF/A-2a / PDF/A-3a 'accessible' conformance levels.
What makes a PDF accessible?
A real tag tree (headings, paragraphs, lists, tables marked as such), a logical reading order, alternative text on meaningful images, a defined document language (e.g. Hebrew), a document title shown in the viewer, and tables with proper header cells. A scanned image with no text layer is the least accessible PDF of all - it is just a picture.
What is the legal risk of an inaccessible document in Israel?
The accessibility duty is anchored in the Equal Rights of Persons with Disabilities regulations (2013), under the Equal Rights of Persons with Disabilities Law. A person with a disability can sue for statutory damages of up to about NIS 50,000 without having to prove actual damage (for websites, subject to prior notice). This is general information - confirm the current figures and your specific obligation with a lawyer.
Does this apply to a foreign company operating in Israel?
If you provide a service to the Israeli public - a customer portal, downloadable forms, invoices, reports - the accessibility duty can apply to your digital documents the same as to a local business. The technical bar is WCAG 2.0 AA, which is an international standard, so a document made accessible to WCAG 2.0 AA already meets the Israeli technical requirement.
How do I make a scanned PDF accessible?
First make it searchable with OCR so it has a real text layer (a scanned image alone can never be accessible), then add tags, reading order and a document language. See our guide on making a scanned PDF searchable.