PDF to DOCX

Convert your PDF files to DOCX, right here in your browser.

Free
Private
No limits
No upload

... or drop files here

Files never leave your device

Add a PDF file to convert.

Frequently asked questions

What is actually inside a DOCX file?

A .docx file is not one document but a ZIP archive of XML parts — one for the text, others for styles, images and relationships. Rename it to .zip and you can open it up and look.

Are my files uploaded to a server?

No. Your PDF is read and the DOCX file is built entirely inside your browser, so nothing ever leaves your device.

Can I convert several files at once?

Yes. Add as many PDF files as you like and download the results one by one, or all together as a ZIP archive.

Will the result look exactly like the original PDF?

The text, headings and table columns are recovered and rebuilt as a DOCX file. This is a content-faithful conversion rather than a copy of the original page design, so fonts, spacing and images are not carried over.

What happens with a scanned PDF?

A scanned page is a picture of text rather than text itself, so there is nothing to extract from it. Convert it to an image format instead, or run it through OCR first.

Is there a file size or file count limit?

No. Everything runs on your own device, so the only practical limit is how much memory your browser has available.

Do I need to install anything?

No. There is nothing to install and no account to create — the converter runs in any modern browser, on desktop and mobile alike.

Why doesn't my Word file look like the PDF?

Because a PDF stores glyph positions rather than a document, so there is no layout to translate — only positions to interpret. What comes across is the text and its heading structure, formatted cleanly; the columns, fonts, colours and page breaks do not. Every PDF-to-Word converter faces this, and the ones that appear to preserve layout do it by filling the page with text boxes, which looks right until you try to edit a sentence.

My converted document is empty. What happened?

Your PDF is almost certainly a scan — a picture of a document rather than a document. There is no text layer to extract, so an empty result is the honest one. Open the PDF and try to select a sentence: if nothing highlights, the file needs OCR before any converter can read it.

Are my headings preserved?

Two levels of them, applied as real Word heading styles so the navigation pane and automatic tables of contents work. They are detected by comparing each line's type size to the document's body size — half again as large becomes Heading 1, about a fifth larger becomes Heading 2. Headings that were marked by boldness rather than size will not be caught.

Do images come across?

No. The conversion reads the text layer, and pictures are separate objects in the PDF. If you want the images, the Extract PDF images tool pulls out the embedded originals at their full resolution, which is better than whatever a document converter would have scaled them to.

What happens to tables?

The text comes through row by row, but not as a Word table object. The underlying column structure is detected — the same detection drives the spreadsheet conversions — so if your document is mostly tabular, PDF to Excel will give you a far more useful result than PDF to Word.

Is my PDF uploaded to a server?

No. It is read and the .docx assembled inside your browser tab. A .docx is a ZIP of XML parts, which can be built in the page with no server involved — so contracts, medical letters and anything under NDA are converted without ever being transmitted.

The hard truth about this conversion is that a PDF does not contain a document. It contains drawing instructions: put this glyph at this coordinate in this font at this size, then the next one, and so on to the bottom of the page. There are no paragraphs in there, no headings, no sentences — those are things your eye assembles from the positions.

So converting a PDF to Word is not a format translation. It is a reconstruction, and every converter that does it is guessing, including the expensive ones. What this tool does is guess carefully and tell you how.

It reads the position, size and content of every text run on the page, groups runs that share a baseline back into lines, and then works out which lines are headings by comparing their type size against the size that carries most of the document's text. A line half again as large as body text becomes a Heading 1; one about a fifth larger becomes a Heading 2. That relative comparison is the trick — it works on a document set in 9pt as well as one set in 12pt, because it never assumes an absolute size means anything.

How to convert PDF to Word

Add your PDF
Drop it in or choose it. The file is read in the page; nothing is uploaded at any point.
Convert
Every page's text runs are extracted with their coordinates, rebuilt into lines, classified by size, and written into a .docx — which is itself a ZIP of XML parts, assembled here in the browser.
Download and check the headings
Open it in Word. Headings will be Heading 1 and Heading 2 styles, so the navigation pane works and a table of contents can be generated — but glance over them, because size-based detection is a heuristic and a document with unusual typography will fool it.

What you get, and what you don't

All the text: In reading order — down the page, then left to right — with runs on the same baseline joined into lines. Sub- and superscripts stay attached to their line rather than breaking it.
Two levels of heading: Applied as real Word heading styles, so the document is navigable and editable as a structured document rather than a wall of paragraphs. Two levels is as far as font size alone can honestly be pushed; a third would be inventing distinctions the geometry does not support.
Editable, selectable text: The point of the exercise. Text comes out as text you can rewrite, not as an image of a document.
The original page design: Not reproduced. Columns, exact fonts, colours, margins, headers, footers and page breaks are not carried over. What you get is the content with its structure, formatted cleanly.
Images: Not extracted into the Word file. To pull pictures out of a PDF, the Extract PDF images tool does that specifically and gives you the embedded originals at full resolution.
Tables: Come through as text, line by line. The column structure is detected — see the spreadsheet targets, which use it — but it is not rebuilt as a Word table object. A table-heavy PDF is better converted to Excel.

The scanned PDF problem

If your PDF came from a scanner or a photograph, it very probably has no text in it at all. Each page is one large image, and the words you can see are pixels arranged in the shape of letters — there is no text layer to extract, and this converter will correctly produce an empty document.

That is not a failure of the tool; it is an accurate report of what is in the file. Getting text out of a scan requires optical character recognition, which is a fundamentally different operation: recognising shapes as characters rather than reading characters that are already there.

The quick test is to open the PDF in any viewer and try to select a sentence with your cursor. If the text highlights, this converter will find it. If you get a selection rectangle over an image instead, you need OCR first.

Where the heading detection goes wrong

It is a heuristic, so it is worth knowing its failure modes rather than being surprised by them.

A document that signals headings by weight rather than size — bold 11pt above regular 11pt — has no size difference to detect, and those headings arrive as ordinary paragraphs. A document with a large decorative pull-quote will have that quote promoted to a heading, because by size it is one. A title page set entirely in large type may come through as a run of headings with no body text beneath them.

All three are quick to correct in Word once you know to look, and all three are the price of not having a document structure to read in the first place.

About the DOCX format

.docx is the word-processing format Microsoft introduced in 2007. It is published as an open specification under the name ECMA-376 / ISO/IEC 29500, so a file can be read without the program that wrote it.

A .docx file is not one document but a ZIP archive of XML parts — one for the text, others for styles, images and relationships. Rename it to .zip and you can open it up and look.

Files like this usually come out of Microsoft Word, Google Docs, and LibreOffice Writer.

What comes through

body text
heading levels
table rows and columns
reading order
bulleted and numbered lists

What does not

the exact page design
embedded images
embedded fonts
form fields

Worth knowing

Because the format is an open, published one rather than a proprietary blob, a {ext} file can be read by LibreOffice, Google Docs, Apple's iWork apps and this page alike — none of them needs Microsoft Office installed to make sense of it.
Heading styles are read as structure rather than as big text, so the levels in the original become real headings rather than a guess based on font size.
A scanned PDF is a picture of a page, not text, so there is nothing for this tool to extract — the result will be empty. That is the honest outcome, not a failure.
This conversion is content-faithful, not pixel-faithful: text, heading levels and table columns survive, the original page design does not.

Information