The tool reads the PDF's text layer with PDF.js, groups characters into lines and lines into paragraphs, and writes a genuine Open XML .docx that Word, Google Docs, LibreOffice and Pages open and edit normally.
You get editable text with its structure intact: headings as headings, lists as numbering, tables as tables, and columns still in columns. You do not get a pixel-perfect clone. Embedded images do not carry over, and the exact font files are replaced by the closest family Word has. A scanned PDF has no text layer and needs OCR first.
The reason to convert a PDF to Word is almost always the same: somebody needs to change the words. Update last year's report, quote a source accurately, translate a letter, rebuild a document whose original file is long gone. The tool above is built for that job, and it is honest about the one it is not built for: reproducing the page exactly as it looked.
How to convert a PDF to Word
Check the PDF has text
Open it in any reader and try to select a sentence. If words highlight, there is a text layer. If the whole page selects as a block, it is a scan and needs OCR before anything can be extracted.
Add the PDF
Drop it or browse. It is read into browser memory; nothing is uploaded.
Press Convert to Word
Every text fragment is read with its position, then grouped into lines, paragraphs, columns and tables, and written as a real .docx in the browser.
Download and open the .docx
Open in Word, Docs or LibreOffice. Headings, lists and tables are real Word objects, so the navigation pane works and restyling the document changes it everywhere at once.
Open XML, the same ZIP-and-XML structure Word writes. Not RTF, not HTML dressed up.
Paragraphs, headings and reading order. Not columns, fonts or images.
The text is extracted and the document assembled in your browser.
What the converter actually does
A PDF page does not contain sentences. It contains instructions like "draw the glyph for T at x=72, y=708". Words exist because glyphs happen to sit next to each other. The converter reads those positioned fragments, sorts them by vertical position into lines, and joins consecutive lines with similar spacing into paragraphs. A gap larger than a normal line break becomes a paragraph break. That is enough to turn a report, a letter or an article into something you can edit like any Word file.
- Body text and its reading orderYes, as paragraphs
- Paragraph breaksYes, from line spacing
- Exact fonts and sizesReplaced with a standard font
- Multi-column layoutKept as Word columns
- TablesCell text as paragraphs, no grid
- Images and figuresNot carried over
- HyperlinksAs plain text
- Scanned pagesNothing; OCR first
Which documents convert well
Running text in one column
- Reports, letters, essays, articles and manuals
- PDFs exported from Word, Docs or LaTeX in the first place
- Contracts and policies with numbered paragraphs
- Any PDF where you can already select the text
Layout carries the meaning
- Magazine and newsletter layouts with several columns
- Forms, where labels and blanks are positioned by hand
- Tables and financial statements: use PDF to Excel
- Slides: use PDF to PowerPoint
- Scans without a text layer: OCR first
Columns are the classic trap. The eye reads down one column and then the next, but the characters' positions only say where they are on the page, not which column they belong to. A two-column newsletter can come out with the columns interleaved line by line. If the document matters and has columns, expect to reorder a few paragraphs in Word.
Getting the cleanest result
- Confirm there is text first. The select-a-sentence test takes five seconds and saves a confusing empty output.
- Single-column sources convert best. If you have a choice of PDF, use the one with the simplest layout.
- Expect tables as text. The cell contents arrive as paragraphs. Rebuild the grid in Word, or convert the table pages with PDF to Excel and paste.
- Style in Word, not here. The output uses one clean font and standard margins on purpose. Applying your own heading styles to plain text is faster than undoing a converter's guesses.
- Watch for hyphenation. Words broken across lines in the PDF may keep their hyphen. A find-and-replace for "- " catches most of them.
Scanned PDFs and OCR
A scan is a photograph of a page. There are no characters in it, only pixels, so this converter has nothing to read and produces an empty document. The fix is optical character recognition, which looks at the pixels and writes a text layer. Run the scan through an OCR tool first, then convert the OCR'd PDF here. The quality of the Word file will match the quality of the OCR, so a clean, straight, well-lit scan pays off twice.
Why the conversion runs in your browser
The PDFs people convert to Word are contracts, medical letters, tax documents and drafts with personal details, precisely because those are the ones that need editing. Most converters upload the file to a server, convert it there and email a link, which hands the full contents to a third party for a format change.
The tool is right above
Everything on this page runs in your browser. Scroll up, drop the file, and keep it on your machine.
Common questions
What people ask with the .docx open in Word.
01Is the Word document actually editable?
Yes. It is a real .docx, and the structure is real too: headings use the built-in Heading styles, lists use Word numbering, and tables are Word tables. Open it in Word, Google Docs, LibreOffice or Pages and edit it like any document.
02Does it keep the layout, fonts and images?
Most of it. Headings become real Word headings, lists become real numbering, tables come through as tables with their merged cells and repeating header rows, and a two-column layout stays two columns. Page size and margins are taken from the source rather than assumed. Two things do not carry over: embedded images, and the exact font files, which are replaced by the closest family Word actually has installed.
03Will it work on a scanned PDF?
Only if the scan already has a text layer from OCR. A pure image scan has nothing to extract. Run it through an OCR tool first, then convert the result.
04Why are my columns mixed up?
They should not be. The converter looks for the whitespace gutter running down the page and reads the left column before the right, rather than taking text in the order the file happens to store it. On the test corpus it gets the order exactly right on every tagged document and every academic paper. If a specific file still comes out interleaved, it is worth reporting, because that is a defect rather than a limit.
05What happens to tables?
A table that the PDF draws with ruling lines becomes a real Word table, including cells merged across columns or rows, and a header row that repeats if the table breaks across a page. A table laid out with spacing alone and no lines drawn is harder to see, and that text arrives as paragraphs. For those, PDF to Excel is often the better route.
06Is there a page limit?
No fixed limit. Every page is read in your browser, so very long documents take longer, but there is no cap.
07Can I convert a password-protected PDF?
Not while it is encrypted. Remove the password from a file you own with the Unlock tool first, then convert.
08Is it free?
Free. Every tool, unlimited use, no signup.