PDF to Word
Get a document you can retype in, not a picture of one.
This tool uploads your file for conversion and deletes the temporary copy when the request finishes. What happens to your files
This tool uploads your file to a server to convert it.
Your file is uploaded to a server for this conversion. It is written to a temporary directory, used, and deleted before the response is sent — within seconds, in the same request. It is never written to a database, an object store, or a queue, and it is never logged.
What survives: Office-made PDFs (Word, Google Docs, LibreOffice exports) convert with real paragraphs, headings and tables. Scanned PDFs are recognised with OCR automatically, which takes longer. PDFs made by print or typesetting software may lose the spacing between words. Left-to-right scripts only.
A quick browser check runs when you start.
How it works
- Read the live upload and retention statement.
- Choose a text-based PDF and start conversion.
- Download the DOCX and compare headings, tables, and columns.
Key benefit
Reconstructs editable paragraphs and tables from positioned PDF content.
Supported files and key limits
Input: PDF. Output: DOCX.
Scanned pages are recognised with OCR (English, German, French and Japanese) and come back as plain text; print-made PDFs convert as a clean editable document rather than a copy of the layout; right-to-left scripts are not supported.
What you get
PDFOak rebuilds real Word paragraphs and tables, with working heading styles, so you can click into a sentence and keep typing instead of editing a pile of floating text boxes. Check the result anyway, and know the three cases. A PDF made by Word, Google Docs or LibreOffice converts with its paragraphs, headings, lists and tables. A PDF typeset for print, such as a journal article or a LaTeX document, converts as a clean editable document that follows the reading order rather than copying the page layout. A scan is recognised with OCR and comes back as plain text.
Why PDFs do not contain paragraphs
A PDF stores instructions to draw glyphs at coordinates. Nothing in it says these forty words are one sentence or that this sentence belongs to that paragraph, so the structure has to be inferred from position, size and spacing. A common shortcut is one floating box per line. It looks right and is miserable to edit, because you cannot reflow anything and deleting a word leaves a hole.
What survives and what does not
Text, reading order, tables, images and most headings come across. Exact visual layout is approximated: text reflows to the page width, spacing is regularized, and a font you have that the server lacks is substituted. Two-column layouts are fine. Brochures and heavily designed pages will need tidying.
The trade is deliberate. People who convert a PDF to Word usually want to change something, such as a figure, a clause or three paragraphs to reuse. Editable content that needs restyling beats a pixel-perfect page you cannot type in. Only left-to-right scripts are supported, so Arabic and Hebrew documents should not be converted here.
Scanned or protected PDFs
A scan holds a picture of text, not text. The converter recognises it with OCR (English, German, French and Japanese) and returns the words as plain paragraphs, so expect to restore the formatting yourself, and expect mistakes where the scan is skewed, faint or handwritten. If you also want a searchable PDF, run OCR PDF on it. The converter also refuses a password-protected PDF. Remove the password with Unlock PDF first.
This one uploads your file
Reading a PDF’s layout well enough to rebuild it takes more than a browser tab can do, so PDFOak uploads the file, converts it, and deletes the temporary copy when the request finishes. The privacy page lists which PDFOak tools upload.