You design a flyer, a report, or an invoice using a distinctive font, export it to PDF, and it looks exactly right on your machine. Then a colleague opens the same file on their laptop and the headline has silently switched to a generic system font, the line spacing has shifted, and text that used to fit on one line now wraps to two, pushing content off the page.
This happens because the font itself was never included inside the PDF file, only referenced by name. If the recipient's device lacks that exact font, the viewer substitutes something close, and "close" is rarely close enough for a letterhead, invoice, or any document with precise pagination.
Embedded versus referenced fonts
A PDF can store fonts two ways. An embedded font ships as actual data inside the file, so any device rendering the document uses the exact letterforms the designer intended. A referenced font just names the typeface and leaves it to the viewing device to supply a matching one from its own system. Everyday documents often get away with this because they use common fonts like Arial, but anything using a licensed or less common typeface is at risk the moment it leaves the machine it was created on.
Subsetting: the practical middle ground
Fully embedding an entire font file can add unnecessary size to a document, especially for typefaces with large character sets. Subsetting solves this by embedding only the specific characters actually used in the document rather than the full font, keeping file size low while still guaranteeing the visible text renders correctly everywhere it is opened.
Why this matters for PDF/A archiving
The PDF/A standard, used for long-term archiving by authorities and financial institutions, requires all fonts to be embedded as a strict rule, precisely because an archived document must render identically decades from now, on hardware that does not exist yet. A file with missing font embedding fails PDF/A validation outright. Converting a document with PDF to PDF/A handles this automatically, embedding any missing fonts as part of producing a compliant archival file.
How to check if fonts are embedded
Most desktop PDF readers list font information under a document properties or fonts panel, showing each typeface used and marking whether it is embedded, embedded subset, or not embedded. Any font marked simply "not embedded" is a risk the moment the file leaves your device, worth checking before sending contracts, official letters, or print-ready designs.
Fixing a document with missing embedding
| Situation | Recommended fix |
|---|---|
| Document needs to stay editable later | Re-export from the source application with font embedding enabled in export settings |
| Only have the PDF, no source file | Flatten with Flatten PDF to lock the exact appearance regardless of fonts |
| Document is going into long-term archive | Convert with PDF to PDF/A to force compliant embedding |
| Need to confirm two versions look identical | Use Compare PDF to catch layout drift between versions |
A short workflow for anything font-sensitive
- Check the font panel in your PDF reader before sending the file out.
- Re-export with embedding turned on if you still have the source file.
- Flatten the document if you only have the finished PDF and no source.
- Convert to PDF/A with PDF to PDF/A for anything needing long-term storage.
- Compare the original and fixed versions with Compare PDF to confirm nothing shifted.
Frequently asked questions
Does embedding fonts make a PDF much larger?
Not usually. Subsetting keeps only the characters actually used in the document, so a typical text-heavy file grows by a modest amount, often under a few hundred kilobytes, far less than embedding the full font library would.
Will flattening a PDF fix font problems without re-exporting?
Yes, flattening converts the text into a fixed visual layer that renders identically everywhere, since it no longer depends on the viewing device having the correct font installed at all.
Why did my PDF/A conversion fail even though the document looked fine?
PDF/A validation checks the underlying file structure, including font embedding, not just visual appearance, so a document can look perfect while still failing compliance because a referenced font was never actually embedded in the data.