Why PDFs lose their formatting when you convert them to Word
Updated · 5 min read
You convert a neat PDF into a Word document and something always goes wrong: the columns collapse, the table turns into loose lines, the fonts change, or every line sits in its own text box. That isn't a broken converter. It's what happens when you turn a format built for printing into one built for editing — and knowing why helps you pick the right approach.
A PDF is a printout, not a document
A Word file knows its own structure: this is a heading, this is a paragraph that flows onto the next page, this is a table with three columns. When you add a sentence, everything after it moves along.
A PDF mostly doesn't know any of that. It records where to draw each piece of text, in which font and at what size — 'put these characters at this spot on the page' — much like a printer would. There is often no paragraph, no table and no column in the file, only text positioned so that it looks like one. Lines don't wrap; they simply stop where the page designer ended them.
So any PDF-to-Word converter has to guess the structure back from positions on the page. Where that guess is easy, the result is good. Where it's hard, the converter has to choose between looking right and being easy to edit.
The two ways converters go wrong
- Copying the look. Some converters place every line or block in its own text box at the exact original position. The page looks almost identical, but editing it is painful: add a word and the line doesn't wrap, the next box doesn't move, and text overlaps.
- Rebuilding the text. Others read the text in order and turn it into ordinary paragraphs. The result is easy to edit, but designed layouts — columns side by side, tables, text placed next to pictures — are simplified.
Neither is wrong; they suit different jobs. If you want to change the wording, rebuilt paragraphs are what you need. If you want an exact replica, Word is the wrong destination.
What converts well and what doesn't
| How it converts | Why | |
|---|---|---|
| Letters, reports, contracts exported from Word or Google Docs | Well | Mostly flowing text in one column, so paragraphs are easy to rebuild |
| Two-column papers and newsletters | Text is fine, layout isn't | The reading order can be found, but the columns become one stream of text |
| Documents with tables | Tables lose their grid | A PDF table is often just text placed in rows, with lines drawn around it |
| Brochures, posters, designed slides | Poorly | Text sits in free positions over pictures, with no paragraph structure |
| Scans and phone photos of pages | Not at all without text recognition | The page is a picture; there are no characters to convert |
Tip: To check for a scan, try to select a word in your PDF viewer. If you can't, or the whole page highlights like a picture, there's no text layer to convert.
What our converter keeps
PDF to Word takes the rebuild approach. It reads each page in reading order — two-column pages one column at a time, not straight across — and creates real, editable paragraphs rather than one text box per line. Lines broken by the page width are joined back together, words hyphenated across lines are re-joined, larger text becomes Word headings so the document outline still works, and each PDF page starts on a new page.
What it doesn't recreate: images, tables, colours, fonts and the exact position of each element. Tables come through as plain lines of text, and pictures are left out — save them separately with Extract Images from PDF if you need them. Scanned PDFs can't be converted yet; you'll see a note that no text was found rather than receive an empty document without explanation.
Get an editable .docx of a PDF's text, made in your browser:
When Word isn't the right target
Often the real goal isn't a Word file at all. Pick the output that matches what you'll do next:
- You only need the words — to quote, search or paste somewhere else. PDF to Text gives a plain .txt file with a '--- Page N ---' line before each page and every line kept as it is on the page.
- You want the structure without the styling — for notes apps, a wiki, GitHub or an AI chat. PDF to Markdown turns larger text into headings, bullet and numbered lines into lists, and re-joins wrapped paragraphs.
- You need to change a few details but keep the design — leave it as a PDF. Filling in a form, signing, adding page numbers or a watermark can all be done to the PDF itself.
- You have the original file — ask for the Word or Google Docs version it was exported from. Going back to the source always beats converting backwards.
Tidying up after converting
- Open the .docx and switch on the navigation pane or outline to check the headings came through.
- Rebuild any tables with Word's own table tool, pasting the converted lines into the cells.
- Apply one set of styles to the whole document instead of fixing paragraphs one by one.
- Add back the pictures you need from the original PDF.
Questions
Why does my converted Word document have text boxes everywhere?
That converter copied the look of each page by placing lines at their exact positions. It looks close to the PDF but is hard to edit. A converter that rebuilds paragraphs, such as PDF to Word, gives text that flows normally.
Can any converter keep a PDF's layout exactly?
A Word document and a PDF lay out pages differently, so complex designs never carry over perfectly in an editable form. If the exact look matters, keep the PDF and edit it directly, or get the original source file.
Why did my PDF convert into an empty document?
It's almost certainly a scan — pictures of pages with no text layer. Without text recognition, there are no words to convert.