Character encoding – Failed
Acrobat’s “Character encoding – Failed” means some text can’t be turned into real characters, so it reads as garbage. How to test it and how to fix it.
Last updated
This message comes from Adobe Acrobat's Accessibility Checker.
Character encoding – Failed is a result from the Accessibility Checker in Adobe Acrobat Pro, in its Page Content category. It means some text in the PDF can't be reliably converted into real characters. The text looks fine on screen, but screen readers, copy and paste, and search get garbled symbols, empty boxes or nothing at all.
What this means
A PDF draws text using glyphs, the shapes in a font. To know which letter each shape stands for, the font needs a map from its glyphs to standard Unicode characters. When that map is missing or wrong, the shape for "f" might be read as a box, a symbol or nothing. PDF/UA, the ISO standard for accessible PDFs, requires every character to map to Unicode, and it doesn't allow text that uses a font's empty placeholder glyph, called .notdef.
Common causes:
- Symbol, icon or decorative fonts that have no Unicode map, such as a check-mark font used for real content.
- Ligatures, like "fi" and "fl" joined together, mapped incorrectly by older software.
- PDFs made by printing to a PDF driver, or by old or unusual tools.
- Text from design apps that was converted in unusual ways during export.
A quick test anyone can do: select a paragraph in the PDF, copy it, and paste it into a plain text editor. If you get garbled characters, boxes or missing letters, that text has an encoding problem. Do it on headings, body text and any special symbols.
Why it matters
Screen readers read the garbled characters aloud, or skip them, so words and sentences come out broken. Braille displays show nonsense. Search inside the document fails, and anyone who copies text into another document gets errors. This affects screen reader and braille display users most, and it's covered by WCAG 2.1 success criteria 1.1.1 Non-text Content and 1.3.1 Info and Relationships.
How to fix it in Acrobat Pro
Adobe's help is candid that Acrobat can't repair some encoding problems. Work through these options, best first:
- Go back to the source file and export again with a current version of the authoring app. Adobe's suggestions include making sure the fonts are installed on your computer, and switching to a different font, preferably OpenType, before re-creating the PDF.
- If you can't re-export, use the copy-and-paste test to find exactly which text is affected.
- For a few short items, such as a symbol, a ligature or a heading in a decorative font, give the tag that holds the text, its label in the document's hidden structure, an Actual Text value. In the Tags panel, right-click the tag, choose Properties and type the correct characters in Actual Text. Screen readers read the actual text instead of the broken glyphs.
- For whole pages of unreadable text, re-creating the document from its source is usually the only practical fix.
- Run the check again and repeat the copy-and-paste test.
Actual Text fixes what people hear, not the font
Actual Text gives assistive technology the right words, but the underlying font problem is still in the file. PDF/UA validators such as PAC may still report it, which is another reason to fix the source when you can.
Fix it in the source file
- Use standard fonts for real text. Use OpenType or TrueType text fonts, and avoid symbol fonts for meaningful content. Type a real character, such as ✓, instead of a letter shown in a symbol font.
- Export, don't print. Use the app's own PDF export, such as File > Save As > PDF in Word or Export > Adobe PDF in InDesign, rather than a PDF print driver.
- Update old templates. If the same text breaks in every document, the template probably uses a problem font.
Fix it automatically with Includoc
This is one of the problems Includoc doesn't fix automatically. Guessing which character a broken glyph is meant to be can put the wrong words into a document, so we flag encoding problems and show you which pages are affected. The fix is to export again from the source. Where that isn't possible, the human-verified tier can add replacement text by hand.
Everything else in the file can still be fixed automatically: tags, reading order, headings, tables, alt text, title and language.
Upload your PDF to fix this automatically
Free check in seconds. Files are deleted within 24 hours.
Standards
| Standard or tool | Reference |
|---|---|
| WCAG 2.1 | |
| PDF/UA-1 (ISO 14289-1) | Clause 7.21.7, Clause 7.21.8 |
| Matterhorn Protocol | Checkpoint 10-001 |
| Acrobat rule | Character encoding |
Related errors
- Fonts are embedded: fonts missing from the file, a related PDF/UA font problem.
- Image-only PDF – Failed: pages with no real text at all.
For more on the standard behind this check, read PDF/UA explained, or check a PDF free to find pages with encoding problems.
Related
- Fonts are embedded
PAC fails “Fonts are embedded” when a PDF uses fonts it doesn’t include. Why PDF/UA requires embedded fonts, how to check, and how to fix it.
- Image-only PDF – Failed
Acrobat’s “Image-only PDF – Failed” means pages are pictures of text, usually scans. Why OCR alone isn’t enough, and how to make a scanned PDF accessible.
- PDF/UA explained: PDF/UA-1, PDF/UA-2 and the Matterhorn Protocol
PDF/UA is the ISO standard for accessible PDF files. What PDF/UA-1 and PDF/UA-2 require, how Matterhorn and validators test them, and the identifier.