PDF to Word
PDF to Word pulls the text out of a PDF, groups it into paragraphs and headings using font size and line spacing, and writes a .docx you can edit. Because a PDF stores positioned glyphs rather than a document structure, complex multi-column layouts, tables and exact styling will not survive. Text-first documents convert well; magazine layouts do not.
1 Add your file
Enter the open password to use it here. It is checked in your browser and never sent anywhere.
2 Choose your options
A PDF does not store a document — it stores glyphs at coordinates. Converting one back to Word is therefore a reconstruction, and it is worth knowing what survives. Paragraphs, headings (detected from font size), reading order and basic emphasis come across well for ordinary single-column documents: reports, letters, papers, contracts. Multi-column magazine layouts, tables, text boxes, footnotes, headers and footers, and exact fonts and spacing do not. If the PDF is a scan with no text layer, nothing comes out at all — run OCR PDF on it first.
How to PDF to Word
Add your PDF
We check for a text layer immediately and tell you if there is not one.
Choose how it should be rebuilt
Heading detection and paragraph joining are on by default and suit most documents.
Download the .docx
Open it in Word, Google Docs, LibreOffice or Pages and edit freely.
Reconstruction, not translation
A PDF stores positioned glyphs, not paragraphs. There are no headings, no tables and no reading order in the file — those concepts existed in the program that made it and were discarded when the page was rendered. A converter has to infer all of them from coordinates, spacing and font sizes.
Each inference is usually right and occasionally wrong, and errors compound. That is why a document can convert almost perfectly for six pages and fall apart on the seventh, and why simple single-column layouts convert far better than magazine spreads.
A worked example
A two-column research paper converts into text that alternates between the columns mid-sentence, because the glyphs were drawn in an order no human reads in. For a document like this, extracting the pages you need and using PDF to Text — which discards layout deliberately, and so cannot get it wrong — is often faster than repairing the converted file.
Limitations worth knowing
- Scanned PDFs contain no text to extract. Run OCR PDF first, and expect conversion quality to be capped by recognition quality.
- Tables without ruling lines are often indistinguishable from aligned text, so the data survives while the grid does not.
- Multi-column layouts, forms and invoices convert poorly because reading order has to be guessed.
- Embedded fonts are usually subsets that cannot be reused, so substitution changes line wrapping and the page count.
- If the original source document exists, use it. Every conversion is a reconstruction; the original is not.
Further reading
Related tools
PDF to Word — frequently asked questions
No, and any tool that promises otherwise is overselling. You get the text, in reading order, split into paragraphs and headings, in an editable document. Exact fonts, spacing, columns and tables do not survive, because the PDF never stored them as such.
A scan has no text layer — it is an image. Run OCR PDF on it first to recognise the words, then convert the searchable PDF it produces.
Not in this version. The conversion is text-first. Use PDF to JPG to pull the images out separately and paste them in where you need them.
By comparing each line's font size against the document's most common size. Noticeably larger lines become Heading 1, 2 or 3 depending on how much larger. It is a heuristic, and it works well on reports and poorly on posters.
Not as Word tables. A table's cells come through as text in reading order, which you will need to reformat. PDF to Excel does a better job when the content really is tabular.
No. Extraction and .docx generation both happen in this browser tab, which is the point of using this rather than a web service for anything confidential.