Fonts are embedded

Fonts are embedded is a PDF/UA check in PAC, the free PDF Accessibility Checker. It fails when a PDF uses a font without including the font itself in the file. PDF/UA, the ISO standard for accessible PDFs, requires every font used to draw text to be embedded, so the text looks the same and reads the same on every device. PAC's font checks also catch technical font errors, such as character widths that don't match the font.

What this means

A PDF can carry its fonts inside the file, either whole or as a subset containing only the characters used. Or it can just name a font and rely on the reader's device to have it. When a font isn't embedded, the viewer substitutes another one. Text can shift, overlap or show the wrong characters, and the text that assistive technology reads may not match what's printed.

People often search for this message with the words "PDF/UA font". Common causes:

• Fonts that can't be embedded. Some font licenses don't allow it. When Word meets one of these with Bitmap text when fonts may not be embedded turned on, it keeps the look by turning that text into a picture, which is worse for screen reader users.
• The basic PDF fonts. Older tools often skip embedding fonts like Helvetica, Times and Courier, because every PDF reader is expected to have them. PDF/UA makes no exception: they must be embedded too.
• Print-to-PDF and old export settings that leave out common system fonts.
• Text added in Acrobat with a font that wasn't embedded.

To check a file yourself, open it in Acrobat and choose File > Properties (Ctrl+D, or Cmd+D on a Mac), then the Fonts tab. Each font should be listed as "Embedded" or "Embedded Subset".

Why it matters

WCAG 2.1 has no success criterion about embedding fonts. This is a PDF/UA requirement, which is why Includoc maps it to PDF/UA rather than to WCAG. It still affects people. If a substitute font can't show a character, or text has been turned into an image, screen reader users and anyone on a device without the font get a less reliable document. And a single non-embedded font keeps a file from a clean PAC report.

How to fix it in Acrobat Pro and re-check in PAC

Technical font errors, such as missing glyphs, wrong widths or invalid encodings, generally can't be repaired in Acrobat. The fix is to export again from the source with a current version of the authoring app.

Fix it in the source file

• Word: use File > Save As > PDF, never a print driver. If a font can't be embedded, switch to one that can. Don't rely on Bitmap text when fonts may not be embedded: it keeps the look by turning text into pictures that screen readers can't read.
• InDesign: InDesign embeds fonts when it exports, as long as each font's license allows it. Replace any font it reports it can't embed.
• Templates: if the same problem keeps coming back, check the fonts in your templates. Fixing a template once fixes every document made from it.

Fix it automatically with Includoc

This is one of the problems Includoc doesn't fix automatically. Adding fonts to a finished PDF can change how its text looks and reads, so we don't do it. We flag fonts that aren't embedded and technical font errors, and the fix is to export again from the source with font embedding turned on.

Everything else in the file, including tags, reading order, headings, tables, alt text, title and language, can still be fixed automatically. Fix the fonts in the source, export a new PDF, and upload that.

Standards

Related errors

• Character encoding – Failed: text that displays correctly but can't be turned into real characters.
• PDF/UA identifier missing: another requirement that comes from PDF/UA rather than WCAG.

To see how PAC's checks relate to the standard, read PDF/UA explained, or check a PDF free to list any fonts that aren't embedded.