Quick answer
PDF translation fails at the format layer, not the language layer. A PDF records where marks sit on a page — it does not record what a paragraph, a table, or a heading is.
What goes wrong when you translate a PDF
- Paragraphs come out in the wrong order
- Tables lose their rows and columns
- Multi-column pages get read straight across, mixing two columns into one sentence
- Text comes out garbled when the file uses embedded fonts with non-standard character mappings
- Headers, footers, and page numbers get spliced into the body text
Why PDF is harder to translate than Word
- A PDF stores page coordinates, not logical structure — reading order has to be inferred
- Text, tables, and vector graphics are stored separately, with nothing linking them
- Tools that extract plain text throw away the layout before translation even starts
- Scanned PDFs have no text at all, only an image plus an optional invisible OCR layer
How to avoid broken PDF translations
- Translate from the original file when you have it — a Word or PowerPoint source keeps its structure and comes back cleaner
- Reconstruct the logical structure before translating, so body text, tables, and captions are handled separately
- Reflow inside the structure layer rather than pasting translated text back at the old coordinates
- If the source is a scan, treat it as a scanned document so OCR runs once — see how to translate a scanned PDF
Related tools
Bottom line
Most PDF translation problems are structure problems wearing a translation costume. Fix how the file is parsed and the "bad translation" usually disappears.