Why Word to PDF Changes Your Layout, and How to Get Reliable Output

A Word document does not contain pages. It contains text and rules, and whichever program opens it works out the lines and page breaks at that moment. PDF does the opposite: it records exactly where each character sits. So converting means freezing one particular layout, and if the software or fonts doing the freezing differ from what you used, you get a different layout.

Fix three things, and most problems go away: export from Word itself, on the computer that has the fonts, with the layout anchored using proper Word features rather than spaces and blank lines.

Reflow: why one small change moves everything

Open word/document.xml inside a DOCX and you will find paragraphs, runs and style references, but no lines or pages. We show this in what is inside a DOCX file.

To display a page, the software measures each word in the current font, fits as many as it can on a line, stacks lines into a page, then moves on. Each decision depends on the one before. So if one line wraps one word sooner on page 2, that paragraph may take an extra line, which pushes a heading to page 3. Its “keep with next” rule then drags the following paragraph along, and a table on page 9 now lands on a different page. A tiny measurement change at the top can have large effects at the end.

That is why the usual complaints look so dramatic: a stray blank page, a table split in two, a heading stranded at the bottom, a signature block alone on the last page.

Cause 1: fonts

This is the most common culprit. Each font has its own character widths, and a substitute font means a different set of measurements, and so different line breaks.

If the machine doing the conversion lacks your font, it silently substitutes another. A document set in a licensed brand font and converted on a colleague’s laptop, or in a cloud service, can reflow all the way through. There are several ways this happens:

• A custom or licensed font that is not on the other machine.
• Newer Microsoft fonts. Cloud-only fonts in Microsoft 365 download on demand, so a machine that is offline or running an older version may not have them.
• Converting outside Microsoft Office. LibreOffice and many web services run on systems without Calibri or Cambria. LibreOffice ships Carlito and Caladea, which are designed to match the widths of those two, and Liberation fonts for Arial and Times New Roman. So layout can hold up when those fonts are present, even though the letter shapes differ. On systems without such substitutes, it will not.
• A font that forbids embedding. Fonts carry licence flags. A restricted font may fail to embed in the PDF, leaving you with a substitution in the output.

To check the result, open the PDF’s Document Properties in a reader and look at the Fonts tab, or run pdffonts. Each font should say “embedded” or show yes in the emb column. A font you did not choose means substitution happened.

The fix: convert on the machine with the fonts. If you must hand a DOCX to someone else, Word for Windows can embed the fonts in the document (File > Options > Save, under “Preserve fidelity when sharing this document”). That makes the file larger, and works only for fonts whose licences allow it.

Cause 2: the layout engine

Different programs implement Word’s layout rules differently. Word’s format has grown for decades, and tables, floating images, text wrapping and compatibility settings are hard to reproduce exactly. Google Docs, Pages and LibreOffice all open DOCX files well for simple documents, and start to diverge as a document gets more complex. So a web converter built on one of these engines can produce a layout that Word would not, even with all the correct fonts.

Word versions differ as well. A document created in an older version may open in compatibility mode, and using it in another version can shift the layout slightly.

Printers can also play a role. Word offers a compatibility option called “Use printer metrics to lay out document”, stored in the file’s settings. Normally, Word uses its own built-in metrics, but with that option on, it asks the current printer driver, and so the layout can change from one machine to another. If you are chasing a document that moves only on certain computers, this is worth checking.

Cause 3: how the document was built

A document held together with manual formatting is fragile, because each trick depends on exact measurements.

Turning on Show/Hide (the ¶ button) shows you what is really there. Runs of blank paragraphs and dots made of spaces are a sign of trouble ahead.

A routine that gives reliable output

1. Open the file in Word on the computer where it was written. Update fields first (Ctrl+A, then F9) so the table of contents and cross references are right.
2. Check every page in Print Layout view, since this is close to what the PDF will contain. Turn off markup: set the Review tab to “No Markup” or the markup may export with it.
3. Use File > Save As or Export, and choose PDF. Avoid printing to a virtual PDF printer if you can. That route tends to lose links, bookmarks and tags.
4. Open Options in the dialog. Look for “Create bookmarks using: Headings”, and “Document structure tags for accessibility”. On Windows, “ISO 19005-1 compliant (PDF/A)” is the option for archiving, and it requires fonts to be embedded.
5. Choose “Standard” rather than “Minimum size” if you plan to print, since minimum size lowers the image resolution.
6. Review the PDF itself. Page count, tables, footers, links. Then check fonts as above.

Word for Mac has its own dialog, and the options differ by version, so go through its settings separately rather than assuming they are the same. Whichever you use, it is the PDF you should proofread as the final document.

If the file came from someone else, and you lack the fonts, ask them for the PDF, or to embed the fonts. Do not proofread a substituted layout and assume it matches what they see.

Before editing and after exporting

Word does not hold the output steady on your behalf. If you add one sentence to page 1 after your last check, re-export and look through the whole file again, since pagination can change downstream. For long documents with strict requirements, section breaks help contain the damage, because each section can restart its layout.

What about our converter?

Our DOCX to PDF tool takes a different approach from Word, and you should know which you are getting. It reads the content of the DOCX and lays it out afresh in your browser, using its own page-breaking rules and the PDF standard fonts. We ran its function on a test letter with a heading, a bold phrase, a photo and a table. The result was one page and 20,309 bytes, with all of those elements present. pdffonts reported this:

name                                 type              encoding         emb sub uni object ID
------------------------------------ ----------------- ---------------- --- --- --- ---------
Helvetica-Bold                       Type 1            WinAnsi          no  no  no       5  0
Helvetica                            Type 1            WinAnsi          no  no  no       4  0

So the structure survives and your fonts do not. Whatever typeface the document used becomes Helvetica, and page breaks land where the tool decides, so page counts can differ from Word. Word headers and footers are not carried over, text boxes become ordinary paragraphs, and images in formats other than PNG or JPEG get a labelled placeholder. The built-in fonts also cover only Western European Latin. For documents in other scripts, the tool points you to your browser’s print dialog instead, since the characters would otherwise come out as question marks.

Use it when you are away from Word and want a clean, readable PDF of a plain document, such as a draft to read on a phone. Do not use it for a designed CV, a brand-sensitive report, or a contract where page references must match the original. For those, use Word, or LibreOffice if you do not have Word.

When layout problems are really size problems

Sometimes the layout survives, and the PDF is too large because Word exported big images. Compress the pictures in Word before exporting (Picture Format > Compress Pictures), since this is cleaner than shrinking the PDF afterwards. If you already have the PDF, read how to compress a PDF first, as some methods will turn your text into images. Compress PDF has a Lossless mode that leaves text alone.

Quick diagnosis