Document structure

What's Actually Inside a Word Document?

Before you can understand why a Word to PDF converter sometimes gets a document wrong, it helps to know what a .docx file actually is under the hood.

Open a .docx file in a text editor and you won't see readable text — you'll see binary. That's because a modern Word document isn't really a single file at all. It's a ZIP archive containing a small filesystem of XML files, a format Microsoft calls Office Open XML (OOXML). Rename any .docx to .zip and you can unzip it to see this for yourself.

The pieces inside the archive

Inside that archive, a few files do most of the work. document.xml holds the actual text and paragraph structure. styles.xml defines every heading style, font, and paragraph format the document uses. A media folder holds embedded images. A web of relationships files ties all of these pieces together — which image goes where, which style applies to which paragraph, where a hyperlink points.

This matters because a Word document isn't really "text with formatting bolted on." The structure and the content are woven together. A heading isn't just big, bold text — it's a paragraph tagged with a "Heading 1" style, which is itself defined elsewhere in the file and might be referenced by a table of contents field somewhere else again.

The structural elements that make a document a document

• Styles and formatting — fonts, sizes, colors, and spacing are usually inherited from named styles, not applied character by character.
• Sections and page setup — margins, orientation, and page size can change partway through a document via section breaks, which is how one document can mix portrait and landscape pages.
• Headers, footers, and page numbering — often different on the first page, odd pages, and even pages, all tracked separately.
• Tables — with their own row heights, merged cells, and nested formatting rules that are notoriously easy for a converter to mishandle.
• Embedded objects — images, charts, and OLE objects (like an embedded spreadsheet) that need to be rendered, not just copied.
• Fields — dynamic content like a table of contents, page count, or last-modified date, computed at render time rather than stored as plain text.

Why this matters when you convert to PDF

A PDF, unlike a .docx, is a fixed-layout format — every piece of text has an exact coordinate on an exact page. So somewhere in the conversion process, all of that flowing, style-dependent, field-computed structure has to be resolved into fixed positions. That resolution step is where cheap or naive converters fall apart: a merged table cell renders wrong, a section break gets ignored and page orientation flips unexpectedly, or a font that isn't installed on the converting server gets silently substituted for something that changes the whole layout.

A reliable Word to PDF converter has to actually understand this structure — parsing the styles, resolving the fields, laying out the tables — rather than guessing at what the document should look like from surface-level text. That's the difference between a PDF that looks like your document and one that's merely a rough approximation of it.