PDF to Word

A PDF to Word converter takes a PDF and gives back a .docx: a document you can open in Word, Pages, LibreOffice or Google Docs and actually edit. It is the tool you want when the text is the point. A contract you have to amend, a report whose paragraphs are going into something else, a form letter that arrived as a PDF and needs three words changed.

The honest version of what happens is worth knowing before you start, because it explains every result you will get. A PDF does not store a document. It stores instructions for painting glyphs at coordinates on a page. There are no paragraphs in the file, no headings, no table structure, often not even spaces: a run of text can be a single letter positioned beside the last one. Everything a word processor needs has to be inferred back out of those positions.

So this page reassembles. Runs sharing a baseline become a line. Lines whose spacing matches the body text become one paragraph, and a wider gap starts a new one. A word broken by a hyphen at a line end is put back together. Then each paragraph is written into the Word document as a paragraph, in reading order, page by page.

What that buys you is editable prose with the structure a reader cares about: paragraph breaks where the original had them, and page breaks where the original paged. What it cannot buy you is layout fidelity. Columns come out in reading order as the PDF declares it. A table becomes its cells as text rather than a Word table. Images are not carried over, because they are not text and this converter deals in text. Any tool that claims otherwise is either doing full layout reconstruction, which is guesswork that goes wrong in ways you will not notice until you print it, or it is sending your document to a server farm to do it.

That last point is the one worth weighing. Almost every PDF to Word service uploads your file. For a menu, who cares. For a contract, a medical letter, a payslip or anything under NDA, you have just put it on somebody else's disk. Here the PDF is opened by pdf.js inside the browser tab, the .docx is built in the same tab, and the download comes out of memory. You can watch the network panel while you convert: nothing leaves.

A scanned PDF is the one case where no converter of this kind can help. If the pages are photographs of paper, there is no text layer to read, and what you would need is optical character recognition rather than conversion. This tool detects that case and says so instead of handing you an empty document.