Translating a PDF while keeping its formatting intact is genuinely difficult — more so than most people expect. The challenge isn't the translation itself; it's that PDF is a fixed-layout format that wasn't designed to have its text replaced. The honest answer is that you can get close, but the degree of formatting preservation depends a lot on the complexity of the original document.
Why PDF Translation Is Harder Than It Looks
A PDF stores text as positioned characters on a page — not as reflowable paragraphs. When you translate text, the word count and character count almost always change. Spanish translations of English text are typically 20-30% longer. German can be longer still. Chinese or Japanese translations are often shorter. Swapping in translated text that's a different length breaks layouts that were designed around the original text length.
Add to this that PDF doesn't natively support complex scripts like Arabic or Hebrew that read right-to-left, that fonts in the original may not contain the characters needed for the target language, and that translated text in a PDF editor often doesn't reflow the same way word processors do — and the complexity becomes clear.
The Most Reliable Approach: Convert, Translate, Export
For documents where both accurate translation and clean formatting matter, the best workflow is: convert the PDF to Word, translate the Word document, then export back to PDF. This gives you a proper word processor for the translation step — text reflows correctly, styles apply consistently, and you have full control over the layout.
The conversion step works best with text-heavy documents that have straightforward layouts — reports, contracts, articles. Complex layouts with multiple columns, text boxes, tables, and embedded graphics will need manual formatting work after conversion, since Word doesn't perfectly reconstruct every PDF layout.
WukongPDF's PDF Converter tool handles the PDF-to-Word step, after which you can translate in Word using DeepL, Google Translate's document upload feature, or manually, then export the translated document back to PDF.
Using Dedicated PDF Translation Tools
Several tools now offer direct PDF translation — you upload the PDF and receive a translated PDF back. DeepL and Google Translate both support PDF file uploads. These tools attempt to preserve the layout while replacing text in place. For simple, text-heavy documents with minimal formatting, the results are often quite good. For complex layouts, tables, or documents with tight column spacing, the formatting frequently breaks.
Translation quality from these tools has improved significantly and is generally accurate for most common language pairs. The main variable is formatting fidelity, not translation accuracy.
What 'Keeping Formatting' Actually Means in Practice
For most translation use cases, "keeping formatting" means: headings look like headings, paragraphs are readable, tables stay as tables, and the document is navigable. It doesn't necessarily mean pixel-perfect reproduction of the original layout, especially when the target language is significantly longer or shorter than the source.
If you need the translated document to be visually identical to the original — same page breaks, same column widths, same text positioning — you're looking at professional desktop publishing work, not a simple file conversion. That's a reasonable requirement for published materials; it's overkill for internal documents or reference translations.
Translating Scanned PDFs
Scanned PDFs need an extra step: OCR first to extract the text, then translation. Without OCR, translation tools see an image with no text to work with. Run the scanned PDF through an OCR tool to get a text-layer PDF or a Word document, then proceed with translation as normal.
For scanned documents, formatting preservation after translation is essentially impossible to guarantee — the scan itself is an image, and rebuilding the layout around translated text requires manual reconstruction. For scanned documents where only the meaning matters (not the appearance), extracting the text via OCR and translating it as plain text is faster and sets more realistic expectations.
Matching the Approach to the Document
Simple text document, formatting matters: use a direct PDF translation tool and review the output. Simple text document, formatting doesn't matter: extract text, translate, distribute as a separate plain-language version. Complex layout, formatting matters: convert to Word, translate, reformat, export to PDF. Scanned document: OCR first, then translate using whichever approach fits the complexity.
No single tool handles all of these equally well. Knowing which case you're in before you start saves time spent trying approaches that won't work for your specific document.