How to edit a PDF in Word (and when it won't work)
Updated:
PDFs are made to look the same everywhere, not to be edited. When you need to change a contract, a CV or a form, converting the PDF to Word is often the fastest route, but the result is not the document the PDF was made from. We converted a test document and looked inside the result, so you know what to expect.
First, check what kind of PDF you have
Open the PDF and try to select a word. If you can, the PDF contains real text and will convert into a document you can edit. If you can't, it is a scan, an image of a page. Converting a scan gives you a document holding that image, not text: turning it into text needs OCR (optical character recognition), which is a different kind of tool.
Convert the PDF to Word
- Open PDF to Word.
- Upload the PDF.
- Download the DOCX and open it in Word, Google Docs or any word processor.
What we found when we tested it
Our test PDF is a one-page report with a heading, four paragraphs and a small three-row table. Converted to Word, the page looked almost the same as the PDF: every word arrived, in the same place, and the table still showed its lines. Underneath, though, the structure is different. Each line of text became its own text box, 34 of them for one page, and the table was not rebuilt as a table but drawn as lines with the text of each cell in its own box. One small shift appeared too: two words in the table heading touched.
| What we checked | Result |
|---|---|
| Text | Every word arrived, in the same position |
| Paragraphs | Each line became a separate text box (34 on the page) |
| Table | Drawn with lines and boxes, not a real Word table |
| Look of the page | Very close to the PDF, with a small overlap in the table heading |
| Scanned version of the same page | One picture, no editable text |
| Conversion time | 1.7 seconds |
What that means for editing
Small corrections work well: changing a name, a date, an amount or a few words inside a line. Longer edits are awkward, because a sentence you lengthen does not flow onto the next line; the next line is a separate box. Tables cannot be sorted or extended like real tables. The same applies to the other formats this tool makes, ODT, RTF and DOC, since they all come from the same conversion.
Better routes for bigger changes
If you need to rewrite whole paragraphs, ask for the original file the PDF was made from: editing that always beats converting the PDF back. If there is no original, use PDF to Text: it gives you the plain text in reading order, which you can paste into a new document and format freely. In our test it kept every paragraph and lined the table up in columns with spaces. For a scanned PDF, you will need OCR software; recent versions of Word, Google Docs and many scanner apps include it.
Tips for a clean result
Fix the content first and the layout last: correct the text, then adjust spacing and page breaks. Fonts that are not installed on our servers are replaced with similar ones, so check headings and anything in an unusual typeface. Keep the original PDF: if the document will be printed or signed, compare the two side by side before you send it.
How we tested
We wrote the one-page test report ourselves, exported it to PDF, and converted it with this site's PDF to Word tool, which uses LibreOffice's PDF import, the same engine our servers run. We then opened the DOCX and counted the text boxes and tables in it, rendered it back to a page to compare the look, and repeated the test with an image-only copy of the page to stand in for a scan. Longer and more complex PDFs produce more boxes, but the same structure.
Further reading: Wikipedia — Optical character recognition (OCR)