Two modes, and how to choose
Editable text reads the words out of the PDF and writes them into a Word document as real paragraphs. You can type into it, spell-check it, change the wording and reuse the text. Headings are detected from the font sizes the document itself uses, so a title stays a title and Word's navigation pane works.
Exact look does something quite different: each page is rendered as a picture and placed full-width in the document. What you get looks exactly like the PDF, prints exactly like the PDF, and cannot be edited at all - it is a page in a Word wrapper.
The choice is simply what you need. Changing two sentences in a letter: editable text. Putting a scanned certificate into a report that has to look right: exact look. There is no mode that gives both, and any converter promising one is overselling what is possible.
Why the layout is not perfect
This is the question everyone has, and it deserves a real answer. A PDF does not record that something is a table, a column, a caption or a bulleted list. It records that a particular character sits at a particular point on the page, in a particular font, at a particular size. The structure you see is something your eyes assemble.
Converting back to Word means rebuilding that structure by inference: pieces of text sitting on the same baseline are a line, lines with consistent spacing and left edges are a paragraph, text noticeably larger than the body is a heading. That reasoning handles ordinary flowing documents well - letters, reports, articles, notes.
It does not rebuild two-column magazine layouts, tables with ruled cells, text wrapped around images, footnotes tied to their references, or headers and footers separated from body text. Those arrive as plain paragraphs in reading order. Desktop software costing real money does better; a browser with no server behind it does not, and we would rather tell you that here than let you find out after an hour of fixing.
Scanned PDFs
A PDF from a scanner or a phone camera holds pictures of pages, not text. Editable text mode will tell you there is nothing to extract rather than hand you an empty document.
Two ways forward. Exact look mode works perfectly well on a scan, because it is putting the picture into Word anyway - which is often exactly what people want when they are assembling a document from scanned paperwork.
Or run OCR PDF first. That recognises the letters in the pictures and produces a PDF with a real text layer, which this tool can then convert to editable text. Recognition is not flawless, so read the result before relying on it.
How the .docx is built
A .docx is a ZIP file containing XML that describes the document, and this tool writes that XML directly. There is no document library doing it, which keeps the download small - only a ZIP writer is needed - and means the file contains nothing but what we put in it.
What goes in is paragraphs, heading styles, page breaks between pages, the page size and orientation taken from the PDF, and in exact-look mode the page pictures. The default font is Calibri at 11 point, which is what Word uses, so the document looks normal when it opens.
The result is a plain, standard document that Word, Google Docs, LibreOffice and Pages all open without complaint. Nothing exotic, nothing proprietary, no repair prompt.
Nothing is uploaded
Files are not uploaded - everything happens in your browser. The PDF is read in this tab and the Word document is built in memory. Contracts, medical letters, statements and school paperwork stay on your device.
This also explains a difference you may notice against online converters that ask you to wait in a queue: there is no queue, because there is no server doing work. The only thing fetched is the code, once, after you choose a file.
Pages are processed one at a time with a progress bar, which keeps a modest phone responsive on a long document.
It is also why there are no size rules beyond what your own device can manage. A hundred-page report converts happily on a laptop; on an old phone the same file will simply take longer, and in exact-look mode it is worth dropping the resolution to 120 dpi so the pictures stay a sensible size.
Nothing is stored between visits either. There is no history, no list of recent files and no cache of your documents - close the tab and the only trace is the file in your downloads folder, where you put it.
The reverse direction, and related tools
Word to PDF converts a .docx the other way, in the same browser-only fashion. It is the more reliable direction of the two, because going to a fixed-layout format loses nothing structural.
If you only need the words with no formatting at all, PDF to TXT is simpler and exact. For a table, PDF to Excel reads it into a spreadsheet. For notes or documentation, PDF to Markdown gives you headings and lists in plain text.
And if the aim is to change a couple of things and send the file back as a PDF, consider whether you need Word at all: our PDF tools can add a signature, black out information, crop pages, add page numbers or a watermark, all without a round trip through another format.