PDF

PDF Made from Word Is Broken: Causes and How to Fix

A PDF exported from Microsoft Word that opens broken is rarely a Word problem in isolation. Either the source Word document had issues that the export carried forward, the export tool was wrong for the job, or the conversion settings produced output that’s technically invalid even if Word handled the source fine. This guide identifies which scenario you’re in and how to fix each one.

Quick fix

  1. Open the source .docx file in Word and check it carefully — does it open cleanly? Are there visible glitches, missing fonts, or broken layout? If yes, fix the source document first; the PDF export will inherit those problems.
  2. If the source looks fine, try Word’s built-in PDF export rather than any third-party tool: File > Export > Create PDF/XPS Document. Avoid “print to PDF” workflows for important documents.
  3. In the export dialog, choose “Standard (publishing online and printing)” rather than “Minimum size”, and confirm the Options dialog does not have “ISO 19005-1 compliant (PDF/A)” enabled unless you specifically need PDF/A — this option produces stricter files that some readers handle poorly.

If the resulting PDF opens cleanly, the original export route was the problem. Move on; document the working method for next time.

If that didn’t work

If Word’s native export still produces a broken PDF, the source document likely contains problematic content — corrupted embedded fonts, broken table structures, malformed equations, or images saved in unusual formats.

Try opening the source .docx in LibreOffice Writer (free, available for Windows, macOS, and Linux). LibreOffice’s PDF export is independent of Word’s, with different tolerances for document quirks. Open the file in Writer and use File > Export As > Export as PDF. A surprising number of “Word can’t make a clean PDF of this” cases produce a working PDF straight from LibreOffice.

If the broken PDF is the only thing you have left and the source .docx is gone, attempt a structural repair with qpdf:

qpdf --linearize input.pdf output.pdf

This rebuilds the cross-reference table, which fixes a common failure mode where Word’s third-party export plugins write malformed structure. See the complete guide to qpdf for full options including the --check diagnostic command.

Advanced recovery

When qpdf can’t repair the file, Ghostscript can re-render it as a new PDF, salvaging the visible content even from severely malformed structure:

gs -sDEVICE=pdfwrite -dPDFSETTINGS=/prepress -dNOPAUSE -dQUIET -dBATCH -sOutputFile=output.pdf input.pdf

This drops form fields, annotations, signatures, and tagged structure. For a Word-derived PDF those losses are often acceptable, since Word documents rarely include those elements in the first place — but it’s worth being explicit about the tradeoff before running it.

If neither qpdf nor Ghostscript salvages the PDF, the only remaining path is to recreate the document from a backup of the source .docx or its draft history. Microsoft 365 keeps version history for files stored in OneDrive and SharePoint; check there before assuming the source is gone.

Why this happens

Word-to-PDF conversion is a more complex operation than it looks. Word stores documents as a ZIP-of-XML structure (the .docx format); a PDF stores documents as a stream of objects keyed by a cross-reference table. The conversion has to translate every Word feature — fonts, tables, equations, embedded objects, headers, footers, change tracking — into the PDF object model. Several things commonly go wrong.

Third-party PDF printers produce non-standard output. Tools like CutePDF, Bullzip, doPDF, and other “print to PDF” virtual printers vary widely in how strictly they follow the PDF specification. Files they produce often open in Acrobat, which is forgiving, but fail in stricter readers like Apple Preview or some browser PDF viewers. The fix is to use Word’s native export instead of the print path.

Embedded fonts don’t subset correctly. When Word exports a PDF, fonts used in the document need to be either embedded in the file or substituted at render time. Subsetting — including only the glyphs actually used — can fail for fonts with unusual encoding or licensing restrictions. The result is a PDF where text displays as boxes, question marks, or wrong characters.

Complex tables and headers trigger export bugs. Heavily-formatted tables (especially nested ones), tables that span page breaks awkwardly, and headers that include floating elements have historically been fragile in Word’s PDF export. A document that displays fine in Word can produce a PDF where the table is missing or rendered as overlapping text.

The source .docx was already damaged. If Word’s .docx itself is corrupt — perhaps from a crashed save, OneDrive sync conflict, or a malformed template — the export will inherit the problem. The PDF can’t be cleaner than its source. This is when fixing the Word document first is the necessary step.

PDF/A compliance produces stricter output. Word’s “ISO 19005-1 compliant” option enforces strict PDF/A rules: no transparency, no encryption, certain font requirements. Files that would otherwise export fine sometimes produce errors with this option enabled. If you don’t need PDF/A specifically (typically for archival or legal compliance), turn it off.

Preventing this in future

The single most useful change is to standardize on Word’s built-in export — File > Export > Create PDF/XPS — and avoid third-party PDF printers for important documents. The native export is updated by Microsoft, tested against modern PDF readers, and produces structurally cleaner files than most alternatives.

For documents with complex formatting, run a quick sanity check on the source before exporting: open it in Word’s reading view, look for layout problems, and confirm all fonts are showing correctly. If anything looks off in Word, it will look off in the PDF too.

For documents that absolutely have to work on every reader and platform — public-facing documents, legal contracts, regulatory filings — consider exporting once and then verifying the output in at least three readers: Acrobat, Chrome’s built-in PDF viewer, and Apple Preview if available. A file that opens in all three is far more likely to open for an arbitrary recipient than one tested only in Acrobat.

If the source Word document is itself broken, the Word repair guide covers the diagnostic and fix path before PDF export becomes relevant. For PDFs broken by other origins — email transit, online merge tools, scanning software — the PDF repair pillar lists the symptom-specific guides for each. And if you’re working with a large batch of PDFs that all came out broken from the same Word template, the underlying template itself is often the culprit; recreating it from scratch in a fresh Word document is faster than fighting with the broken one.

Last verified: April 2026