← All posts

Why won’t this tool accept my .docx file?

You have a Word document. You found a page called "Word to PDF". You uploaded the file and it came back with an error, or a red outline, or — worst of all — nothing at all, as if the upload had never happened.

So you tried another converter. Same. And now you are in the third one, hunting for a setting you suspect does not exist, wondering whether the problem is the website or the file.

It is one or the other, and they look identical from where you are sitting. Ten seconds of checking tells you which, and after that the fix is short in both cases.

There are two different refusals here

The tool never accepted .docx in the first place. Its name says "Word to PDF" because that is what people search for, not because a Word document is what it takes. No amount of retrying will help, and no setting will change it.

The tool does accept .docx, but not yours. Something about your particular file fails the check at the door. This is the interesting case, because the file usually opens perfectly well in Word, which is why nobody believes the error.

Ours is the first kind, and we would rather say so on this page

Word to PDF does not convert .docx. Not on a good day, not with a different setting, not if you try again tomorrow.

The reason is worth one paragraph, because it is the same reason behind every converter that quietly mangles your document instead. Laying out a Word file faithfully means implementing Word's layout engine: styles, numbering, section breaks, floating objects, tables, fields, and the line-breaking rules that decide where page seven ends. Nobody writes that from scratch. Every service that offers .docx to PDF is running LibreOffice on a server and handing you what it produced. LibreOffice is not installed on ours, so the tool refuses the file and tells you what to use instead. Nearly right, in a document, is worse than a clear no — you find the wrong bit after you have sent it.

The refusal is done on the file's contents, not its name, so renaming will not sneak one past. It reads the first four bytes. If they are PK\x03\x04 or D0 CF 11 E0, it stops and says what those signatures mean. What it does convert is plain text, Markdown and HTML, which is a genuinely useful thing if the words are what you need.

There is a case where our converter is the right one, and it is more common than it sounds: you do not need Word's layout, you need a clean, readable PDF of the content. A set of release notes. A policy someone wrote in Word for no particular reason. A document you are going to reformat anyway.

Save it out of Word as plain text or Markdown, or write it in Markdown to begin with, and Word to PDF will flow it onto real A4 or Letter pages with proper headings, in one of four typefaces, at a text size and margin you choose. It takes one file at a time, so if you have a README, a CHANGELOG and a licence to combine, convert each and join them with Merge PDF, which can also drop a bookmark at the start of each one. What survives that trip and what does not is worth reading first if your document has images or badges in it, because they do not travel.

But that is the minority case. Most people asking this question have a real Word document with real layout in it and need that layout to survive. The rest of this page is for them — starting with what a .docx file actually is, because that is what most converters are checking when they refuse one.

What a .docx actually is, and why the check is so brutal

A .docx is a zip file. Rename one to .zip, open it, and you will find a folder structure with word/document.xml inside: the text, the styles and the numbering, all as XML, all compressed. This is not a trick, it is the format. Every zip file on earth starts with the two bytes PK — the initials of Phil Katz, who wrote the original PKZIP.

That two-byte signature is the whole of the check most converters run. It is why the error message is always so unhelpful: the tool did not read your document and decide it was bad. It looked at four bytes and stopped.

Six reasons a converter that does take Word files bounces yours

Sorted by how often it turns out to be the answer.

  1. It is a .doc wearing a .docx name. Somebody renamed it, or a system renamed it in transit. The old Word format is not a zip at all — it is an OLE2 compound file, which starts D0 CF 11 E0. Word opens both without comment, so you have no reason to suspect anything. Renaming a file changes nothing about its contents; it only changes what programs guess.
  2. The document has a password on it. This is the one that wastes the most time, because the file is perfectly valid and opens fine once you type the password. When Word encrypts a .docx, it does not encrypt the zip — it puts the encrypted zip inside an OLE2 container. The first bytes are no longer PK. To every converter on the internet your Word document has stopped being a Word document. Remove the password in Word first (File, Info, Protect Document, Encrypt with Password, delete the password, save), then convert.
  3. The local copy is a placeholder, not a file. OneDrive Files On-Demand, Dropbox Smart Sync and iCloud's "optimise storage" all leave a stub on disk where the document appears to be. Explorer and Finder show the real size; the browser uploads the stub. If the file shows a cloud icon rather than a tick, open it once to force a full download, then upload.
  4. The download was cut short. A zip keeps its index — the central directory — at the end of the file, not the start. So a .docx that arrived 99% complete fails exactly as hard as one that arrived 5% complete, and it fails with no useful message. If the file came from webmail, a portal, or a flaky connection, download it again before you debug anything else. Compare the byte count against the source if you can.
  5. The browser told the truth about the wrong thing. Upload forms often whitelist by MIME type, and on Windows the browser gets that type from the registry entry for .docx. If some other program has claimed the extension, the browser announces application/octet-stream and the whitelist bounces a perfectly good file. Symptom: the same document uploads fine from a phone or a different machine. Workaround: use a converter that checks the bytes rather than the label, or fix the file association.
  6. It is not the file you think you picked. ~$report.docx is Word's owner-lock file, a few hundred bytes long, and it sits in the same folder with a near-identical name. Also on this list: .docm and .dotx, which many extension whitelists forget; .odt and .pages, which are also zips and so get past the first check before failing later; and free-tier size caps, which are frequently reported as "invalid file" rather than "too big".

What to check, in order

  1. Look at the size. Under about 10 KB for a document with real content means a stub, a lock file or a truncated download. Stop and get the file again.
  2. Copy it and rename the copy to .zip. If your file manager opens it and there is a word folder inside, you have a real, uncorrupted .docx and the problem is at the tool's end. If it refuses to open, the file is a .doc, encrypted, or damaged — go back to causes 1, 2 and 4.
  3. Open it in Word and use File, Save As. Choose "Word Document (*.docx)" explicitly, to a local folder, not a synced one. That one step fixes causes 1, 2, 3 and 4 at once, because you are writing a fresh, complete, unencrypted file from the program that understands it best.
  4. Then convert it with something that has a real layout engine behind it. Three free options, below.

The three free routes that actually work

Any of these will beat any web converter, because they are the thing the web converters are calling.

Whichever you use, it is worth a look at what came out before you send it. Our PDF inspector reads the page count, the page sizes and whether the text is real text, all in the browser without uploading the file anywhere.

One thing to expect from any of these routes

Running headers, footers and page numbers that a word processor draws for you are properties of the document, not characters in it, and how much survives a conversion depends entirely on where they were put in the first place. If your PDF comes out with none, that has its own short explanation — and its own fix, applied to the finished PDF rather than the source.

The short version: a converter that refuses your file is being more honest than one that accepts it and returns something nearly right. Find out which of the two refusals you have hit, spend the ten seconds renaming a copy to .zip, and then use the program that made the document to make the PDF.

Tools this post uses: