Most PDF-to-Word converters produce a usable Word document. Few produce a clean one. The result usually has the right text but the wrong structure — paragraphs become one giant run, tables become tab-separated lines, headings become bold text without a heading style.
The reason is that PDF and Word model a document very differently. PDF is a rendering format. It stores where to put ink on the page. Word is a logical format. It stores what the content is. Going from PDF to Word means inferring the logical structure from the rendered output — a hard problem.
What usually breaks
- Headings become bold text. The visual cue (larger font, weight) carries over, but the Word heading style doesn't. Navigation pane, outline view, and accessibility tools all break.
- Tables become paragraphs with tabs between cells. The columns shift when you change the font. Sorting, filtering, and formulas don't work.
- Lists become paragraphs with bullets or numbers as text characters. Indentation breaks.
- Footnotes and endnotes become inline text in brackets. Reference markers don't link back.
- Page breaks become hard line breaks. Editing one paragraph can push content onto a different page than expected.
- Images stay in roughly the right place, but text flows around them differently when the font changes.
What DocShop does differently
The PDF to Word tool extracts the document's structure, not just its appearance. Paragraphs stay as paragraphs. Headings become Word heading styles (so the navigation pane works). Tables become real Word tables (so sorting and formulas work). Lists stay as lists.
The conversion uses pdf2docx on the worker — a library specifically built for structural extraction rather than OCR-then-rebuild. The output is editable, not just a flattened image of each page.
What still won't work perfectly
- Scanned PDFs (image-only, no text layer) — DocShop's converter does not run OCR in this version, so a scanned PDF will produce an empty Word document. If you need an editable result, run the scan through an OCR service first (DocuSign, Adobe, macOS Preview, or any free OCR tool), then upload the OCR'd PDF.
- Complex multi-column layouts — newspaper-style columns come through as separate sections, not as flowing columns.
- Forms with embedded scripts — Word can't run PDF JavaScript, so interactive form fields become static text.
- Highly designed documents — brochures, magazines, anything with overlapping elements — Word's structure model can't represent them. The result is a flattened approximation.
For these edge cases, the right move is often PDF to Markdown instead. Markdown's structure model is simpler (headings, paragraphs, lists, code, tables) and easier to maintain than Word for content that's going into a blog or wiki.
When to use PDF to Word vs PDF to Markdown
- PDF to Word when the recipient edits in Word — a contract to redline, a report to revise, a form to fill in.
- PDF to Markdown when the content goes into a blog, README, wiki, or static site. Markdown is simpler, easier to version-control, and converts to Word later if needed via Markdown to Word.
- PDF to Excel when the document has tables you need to sum, filter, or pivot — invoice line items, financial reports, expense data.
The PDF to Excel tool uses pdfplumber for table detection. Detected tables become real Excel sheets with proper columns. Pages without a table become rows of text on a fallback sheet.
Privacy
PDF to Word runs on the same isolated worker as every DocShop conversion. The upload and the result are deleted within an hour. There's no archive, no human access, and no training on your data. This applies to every format conversion, not just Word.
Free tier is 3 conversions per day, 50MB max file size. Pro tier is 100 per month, 200MB. Max tier is unlimited, 500MB.