PDF

PDF Fonts Missing or Display Incorrectly: Diagnosis and Fixes

A PDF that displays text as boxes, question marks, or in the wrong typeface is usually a font embedding problem rather than file corruption. The text is intact in the file; your reader just doesn’t have the fonts the document expects, and the substitution it falls back to either fails or produces visibly wrong results. Most cases resolve in under a minute by switching readers or adjusting one Acrobat preference.

Quick fix

Open the PDF in a different reader. Browser PDF viewers (Chrome, Firefox, Edge) use FreeType for text rendering, which behaves differently from Adobe’s font handling and often succeeds where Acrobat substitutes badly.

  1. Right-click the PDF and choose “Open with…” → your browser of choice.
  2. If the browser also fails, try Apple Preview (macOS) or Sumatra PDF (Windows).
  3. If text now displays correctly, save a fresh copy from the working reader. The browser’s “Print to PDF” option produces a clean output with the visible fonts embedded.

If text displays correctly in any reader, the problem is reader-side rather than file-side, and you’ve already solved it.

If that didn’t work

If multiple readers fail in the same way, the file is missing fonts that aren’t on your system and substitution isn’t producing usable results. Diagnose what’s actually embedded:

pdffonts input.pdf

pdffonts ships with poppler-utils. Install it with brew install poppler on macOS, apt install poppler-utils on Debian or Ubuntu, or download poppler binaries on Windows. Output looks like:

name                   type        encoding  emb sub uni
---------------------- ----------- --------- --- --- ---
ArialMT                TrueType    WinAnsi   no  no  yes
TimesNewRomanPS        TrueType    WinAnsi   no  no  yes

The emb column is the answer. no means the font is referenced by name but not included in the file — the reader has to substitute. If the font isn’t installed on your system, you get boxes, question marks, or whichever fallback your reader picked.

If the source document is available, the cleanest fix is to re-export with embedding enabled. In Word: File → Save As → Tools (or Options) → Save → “Embed fonts in the file”. In InDesign and most design tools, the PDF export panel has a “Subset fonts when percent of characters used is less than” option — set this to 100 to force full embedding, or check “Embed all fonts” if available.

If you don’t have the source, install the missing fonts on your system. Where you find them depends on the font: many common fonts (Arial, Times New Roman, Calibri) ship with Office; design fonts may need licensing; some are freely available from Google Fonts or vendor sites.

Adobe Acrobat has a “Use local fonts” toggle at Edit → Preferences → Page Display. Disabling this forces Acrobat to use only embedded and Multiple Master fallback fonts rather than substituting from your system. The result is often more visibly broken but truer to the file’s actual state — useful for diagnosing whether the displayed text is real or synthesized.

Advanced recovery

If the fonts are unrecoverable and you need a usable file, re-render the PDF with Ghostscript and force font embedding from your system fonts:

gs -sDEVICE=pdfwrite -dPDFSETTINGS=/prepress -dEmbedAllFonts=true \
   -dSubsetFonts=true -o output.pdf input.pdf

What you lose: the original font substitution decisions are baked in. If Ghostscript decides Arial Black should substitute for Helvetica Black, the output PDF will show Arial Black even on systems that have Helvetica Black installed. The decision is no longer the reader’s to make.

If specific glyph corruption is the problem rather than missing embedding — characters appearing as garbled symbols rather than boxes — Ghostscript re-rendering rarely helps, because it preserves whatever character codes are in the file. The fix in that case is to obtain a fresh source file or accept partial recovery.

Acrobat also maintains a font cache that can become corrupted, producing display problems even when fonts are embedded correctly. The cache location varies by Acrobat version and operating system — Adobe’s current support documentation lists the exact paths for your version. Clearing the cache and restarting Acrobat resolves a small but real category of font display issues.

Why this happens

PDFs reference fonts in two ways. Embedded fonts include the font program directly in the file — the PDF carries everything needed to render the glyphs regardless of what fonts are installed on the viewer’s system. Referenced fonts include only the font’s name, with the assumption that the viewer’s system will have it. When it doesn’t, the reader substitutes.

Embedding can be full or subset. A full embed includes every glyph in the font; a subset embed includes only the glyphs the document actually uses. Subsetting keeps file sizes down and is the common default. The hidden cost is fragility: if the document is later edited and new characters are typed that weren’t in the original subset, those characters can’t be rendered from the embedded font and fall back to substitution or display as nothing.

When substitution happens, the reader picks the closest match it can find. Adobe products historically used Multiple Master fonts (Adobe Sans MM and Adobe Serif MM) that could be morphologically adjusted to match metrics of the missing font. Modern readers use various strategies — system font matching by name, by metrics, or simple fallback to a default. The visible result depends entirely on which strategy your specific reader uses.

The text-as-boxes outcome happens when the reader can’t find any reasonable substitute for a particular character. The character code in the PDF maps to a glyph the substitute font doesn’t have — for example, a PDF using a custom symbol font where the substitute is a standard Latin font. The reader renders a placeholder box rather than guess.

PDF/A — the archival PDF subset standardized as ISO 19005 — requires all fonts to be embedded. Files conforming to PDF/A don’t have this class of problem, by design. Files exported as standard PDF have no such guarantee.

Preventing this in future

When you control PDF generation, enable full font embedding in your export settings. The file size cost is modest (typically a few hundred KB per font for full embed) and the resulting file is portable across any system. For documents that will be archived or distributed beyond your control, consider exporting as PDF/A specifically — the format prevents this entire problem class.

When receiving PDFs from others, install commonly-needed fonts proactively. The fonts most often missing from default system installations are Adobe-licensed fonts (Garamond Premier, Minion, Myriad), classic Microsoft fonts on non-Microsoft systems, and design industry standards like Helvetica on systems without Adobe products.

If your workflow involves repeatedly opening PDFs from a particular source — a publisher, vendor, or organization — and font problems recur, ask the source to enable embedding in their export pipeline. The fix is at their end, not yours.

If the file’s images are also missing or broken, the underlying problem is more likely a damaged content stream than a font-only issue — see the guide on PDF images missing or displaying as placeholders for the diagnosis path. PDFs that open completely blank rather than displaying with font substitution failures are addressed in the guide on PDFs that open blank or black. For broader background on PDF structure and the difference between structural and content damage, the PDF repair complete guide covers how PDF files are built and where corruption typically originates.

Last verified: April 2026