Converting a PDF to Word is the most-searched PDF task there is, and the most disappointing. The file opens and the paragraphs are in text boxes, or the columns have merged, or the headings are body text. That is not a bug in a particular converter. It is what the format hides.
Why it is hard
A Word document knows what a paragraph is. A PDF does not. A PDF page is a list of drawing instructions — put these characters at this position in this font — with no notion of a paragraph, a heading, a list, or the reading order between two columns. A word can be drawn as three separate runs because the kerning changed. A table is just text sitting near some lines.
Every converter therefore infers structure: this run of lines is a paragraph because the spacing is regular; this line is a heading because it is larger and followed by a gap; these words are a table because they line up with ruled lines. The inference is good on simple documents and increasingly wrong as the design gets cleverer. The honest converter says so.
What PDF to Word gives you
PDF to Word lists two choices, but it currently runs only one of them. Whichever you pick, you get Looks like the PDF.
“Looks like the PDF”
Each horizontal line of text is placed in the Word document exactly where it sits on the page, with its size, bold, italic and font family, and the page’s images are placed at their positions. Each PDF page starts a new Word page. Open it and it looks like the original.
Choose this when the document has to look the same — a form to sign and return, a letter to reprint on letterhead, a contract where the pagination matters — and you only need to change a few words.
What it is not: flowing text. Insert a sentence and the line does not push the next one down, because each line is anchored to its spot. Paragraphs, headings, lists and tables are not rebuilt, and text printed sideways, such as a rotated table, is left out.
“Editable text”
This choice is listed on the page, but the converter does not use it: the file you download is the Looks like the PDF one, with every line in its own frame. Nothing is rebuilt into paragraphs, headings, lists or Word tables. If you plan to rewrite a document rather than amend it, expect to re-flow the paragraphs by hand in Word.
| You need to… | Use |
|---|---|
| Sign, stamp, or fix a typo and send it back | PDF to Word |
| Keep exact pagination | PDF to Word |
| Reuse the text in a new document | PDF to Word, then re-flow the paragraphs by hand |
| Edit a long report with headings and lists | PDF to Word, then rebuild the headings and lists in Word |
| Convert a scan | OCR first, then PDF to Word |
Step by step
- If the PDF is a scan, run OCR PDF first so there is text to rebuild.
- Open PDF to Word and choose the file. It is read in your browser; nothing is uploaded.
- Leave Looks like the PDF selected; the Editable text choice gives the same file.
- Save the
.docxand open it in Word, LibreOffice or Google Docs. - Check the things converters get wrong: fonts (substituted for standard equivalents — select all and set your preferred font once), tables (they come across as separate lines, not Word tables, so compare a few rows against the PDF), multi-column pages (each line stays in its column, so text will not flow from one column to the next as you edit), sideways text (a rotated table or margin label is left out), text colour and drawn rules (not carried over: a coloured heading comes through black and decorative lines are missing).
When the result will disappoint, and what to do instead
- Magazine and brochure layouts — text wrapped around images, multiple columns of different widths — do not survive any converter as editable text. Use Looks like the PDF, or ask for the source file.
- Forms with fields are better handled by Fill PDF forms than by converting.
- Presentations exported to PDF are a bag of positioned text boxes; Looks like the PDF is the one that makes sense.
- You need the original Word file — a converter reconstructs a document that produces the same page; it cannot recover the file that produced the PDF.
The reverse direction is the easy one: Word to PDF has all the structure it needs.