Convert PDF to an Editable Word Document (DOCX)
Turn a notification, notice or form PDF into a .docx with fonts, colours, tables and images kept in place
Drop a PDF here
Runs entirely in your browser: the text, its styling and its position on the page are read directly from the PDF and rebuilt as a Word document. Bold, italic, text colour, font size and font name all carry over, each page keeps its own size, wrapped lines are joined back into paragraphs, tables with visible borders come through as real Word tables, and side-by-side columns are read one at a time. Photos, logos and backgrounds are pinned to the spot on the page they were drawn at, behind the text, so writing set over a dark panel stays readable, and every block of text is pinned to the position it had in the PDF, so the page comes out looking like the original. A font your computer does not have is swapped for the nearest common one. Borderless tables are not rebuilt, and a scanned page is kept as a picture of the page, since making its words editable would need OCR.
About this tool
A recruitment notice has to be edited into a study note, or a form issued only as a PDF has to be filled in. This tool reads the text of a PDF, with its styling and its exact position on the page, and rebuilds it as a Word document you can edit.
Bold, italic, colour, font size and font name carry over. Each page keeps its size and orientation, wrapped lines are joined back into paragraphs, bordered tables become Word tables, and photos and colour panels sit behind the text where they were.
Conversion runs in your browser; the PDF is never uploaded and there are no options to set. A scanned PDF is a photo of a page, and this tool does not do OCR, so those pages come through as pictures with a notice saying so.
How to convert a PDF to Word
- Drop the PDF onto the tool or click to choose it. Conversion starts at once.
- Wait while the status reads Reading page X of N.
- Read the notice under the download if some pages were scans; those words are not editable.
- Click Download to save the .docx, which carries the PDF's own file name.
- Open it in Word or LibreOffice Writer and edit. Use Choose a different file to convert another.
What converts well and what does not
| Element | Result | Detail |
|---|---|---|
| Running text | Good; wrapped lines are rejoined into paragraphs | Headings and short Label: value rows stay as their own paragraphs |
| Bold and italic | Carried over | Read from the embedded font's name, such as Poppins-Bold |
| Font family and size | Size exact; family mapped | Common fonts keep their name; other sans faces become Arial, serifs Times New Roman, monospace Courier New |
| Text colour | Carried over | Non-black fill colours are applied; black stays the default |
| Page size and orientation | Kept page by page | Each PDF page becomes one Word section with half-inch margins |
| Bordered tables | Rebuilt as Word tables with exact row heights and column widths | Needs a fully ruled grid of at least two rows and two columns |
| Borderless tables | Not rebuilt | Text stays as positioned blocks; nothing on the page marks the cell edges |
| Two to four side-by-side columns | Read column by column | Placed where they were; copy-paste order follows the columns |
| Photos, logos, colour panels, charts | One picture per page behind the text | Rendered at double resolution as a JPEG; not individually movable in Word |
| Scanned pages | A picture of the page | No OCR; the notice counts how many pages were scans |
| Form fields and comments | Left out | Only the printed page content is read |
| Underline, hyperlinks, list numbering | Not rebuilt as Word formatting | Bullet characters and numbers come through as plain text |
How the document is rebuilt
Text arrives from the PDF as small pieces with a font, a size and a position. Pieces on one baseline form a line, unless a gap wider than two and a half character heights separates them. Lines merge into a paragraph when one reaches at least 80 to 90 percent of its column's right edge and the next has the same size and left margin and sits directly beneath.
Each paragraph becomes a Word text frame anchored to the page where its first line sat, with the PDF's own line spacing, which is why the page looks like the original rather than reflowed text.
Hypothetical example: on a 595.28 x 841.89 pt page, a one-line 14 pt heading has its baseline 100 pt from the top and starts 72 pt from the left. Leading is 14 x 1.2 = 16.8 pt. The frame's top is 100 - 16.8 + 14 x 0.22 = 86.28 pt, written as 1726 twips (1 pt = 20 twips); its left is 72 x 20 = 1440 twips; the run's size is written as 28 half-points. The page becomes a portrait section of 11906 x 16838 twips.
Anything that is not text, such as photographs, filled shapes and colour panels, is rendered once per page without the text, saved as a JPEG and pinned behind the text at full page size. Pages with only text and table rules skip this. Tables are recognised from the ruled lines, a stack of horizontals crossed by verticals running the grid's full height.
Scanned PDFs and the OCR limit
A page counts as a scan when a single image covers at least 80 percent of it and no visible text remains. It is kept as that image, at the right size, and the download panel reports a count such as 3 of 12 pages are scans. Making those words editable needs optical character recognition, which this tool does not perform.
Some scans carry an invisible text layer from an earlier OCR pass, drawn in a mode that renders nothing. That hidden text is dropped rather than made visible, since it would land on top of the picture it describes. A PDF with no text and nothing drawn on any page stops with the message that nothing could be read.
Editing the result and what to expect
Because every block is a positioned frame, adding several lines to one block does not push the blocks below it down; the frame grows and may overlap its neighbour. Light edits are fine. For heavy rewriting, select all, copy and paste into a blank document to get flowing paragraphs.
Embedded fonts are not shipped inside the .docx. A font from the common list keeps its name; anything else is replaced, so line breaks can shift by a word or two. The output targets Microsoft Word; LibreOffice Writer opens it as well. A password-protected PDF cannot be read; save a copy without the password first.
Hindi and other Indian-language PDFs convert correctly only when the PDF uses Unicode fonts. Files set in legacy fonts such as Kruti Dev store Latin codes and come out as unreadable characters in any extractor, including this one.
Frequently asked questions
›Can I convert a scanned PDF to an editable Word document?
Not into editable text. A scanned page is kept as a picture of the page inside the .docx, sized to the original, and the download panel tells you how many pages were scans. Turning the picture into editable words needs OCR, which this converter does not include. Run the scan through an OCR tool first, or retype the few lines you need.
›Why does my Word file have text boxes instead of normal paragraphs?
Each paragraph is placed in a frame pinned to the spot it occupied in the PDF, which is what keeps the layout faithful, including columns and text over colour panels. Frames can be dragged and edited in Word like any text. If you would rather have ordinary flowing text, select all, copy, and paste into a new blank document.
›Will the tables from the PDF become real Word tables?
Tables drawn with visible borders on all cells do. The tool reads the ruled lines, finds grids of at least two rows and two columns, and rebuilds them as Word tables with the same column widths and row heights, placing each text piece in its cell. A table set without borders, or with only a few lines, stays as positioned text.
›Why did the font change after conversion?
The PDF carries its own copy of each font, and a .docx does not. Fonts that Word or most computers have, such as Arial, Calibri, Times New Roman, Poppins or Roboto, keep their names. Any other font is swapped for the nearest common face by family, so a sans design becomes Arial and a serif becomes Times New Roman, rather than Word's own fallback to a serif.
›Do I need Microsoft Word installed to convert a PDF to Word?
No. The conversion builds the .docx itself and does not use Word, Acrobat or any installed program. You only need something to open the result afterwards, which can be Word, LibreOffice Writer or any editor that reads .docx. Nothing is installed and there is no sign-up or page limit.
›Does PDF to Word work for Hindi PDFs?
It depends on how the PDF stores its Hindi. A PDF made with Unicode Devanagari fonts extracts correctly, with the text editable in Word. Older files set in legacy fonts like Kruti Dev or DevLys store Latin letter codes under a Hindi-looking font, and every extractor returns gibberish for them; those need the legacy font installed and a converter for that encoding.
›Why does it say nothing could be read from this PDF?
The file had no text and nothing drawn on any page, meaning no images, no filled shapes and no shading. That happens with a PDF whose pages are blank placeholders, or with a damaged export. Open it in a viewer; if the pages look fine there, print it to a new PDF from the viewer and convert that copy.