wxrks doesn't parse a PDF's own file format for translation. Instead, when you upload a PDF as a project's source file, wxrks automatically converts it into an editable DOCX file behind the scenes, then hands that converted file to your linguists in the standard wxrks Editor.
​

💡 Who is this for? This guide is for Project Managers and Account Admins who need to upload a PDF as translatable source content and understand what wxrks does with it within the wxrks platform.

PDF isn't an editable format on its own — there's no reliable way to write translated text back into a PDF's original layout. To get around this, wxrks converts every PDF you upload into a DOCX file the moment it finishes uploading. That DOCX becomes the actual Work Unit (the individual file, inside a project, that a linguist opens and translates) — same Editor, same workflow steps, same everything as if you'd uploaded a Word document from the start.
​

This has one direct consequence worth knowing up front: your deliverable comes back as a DOCX, not a PDF. If you need the finished translation back in PDF form, you'll need to convert the delivered DOCX yourself (for example, by re-exporting it to PDF from Word) after translation completes — wxrks doesn't convert it back automatically.

You upload a PDF the same way as any other source file, as part of creating a translation project: on the Upload Files step, drag your PDF in or click Browse to select it.
​

As soon as the file finishes uploading, wxrks flags it for conversion — click the file's row to expand it and you'll see the notice confirming what's about to happen:

Conversion runs in the background as soon as you upload the file — you don't need to trigger it or wait for it before continuing through the rest of project creation.

By the time you reach the Workflows step, the file already shows its converted name — wxrks keeps the original PDF's full filename and appends .docx to it (for example, a file named Document.pdf becomes Document.pdf.docx), so you can still tell at a glance that it started life as a PDF.

The same converted file shows up under the project's Files tab once the project is created, listed under whichever workflow step it's assigned to:

Because a PDF's layout is being rebuilt from scratch during conversion, it's common to include a DTP (desktop publishing) workflow step for PDF-sourced files — it gives someone a chance to check and fix formatting that didn't survive the conversion cleanly before the file goes out for translation.

Once conversion finishes, wxrks moves your original PDF into the project's Reference Files tab, under the Reference Files section — kept for anyone on the project to open or download for context, but no longer part of the translation workflow itself.

Don't confuse this with the separate PDF Preview Guide Files section on the same tab — that one is for a completely different use case: uploading a PDF rendition of an InDesign (IDML) file purely so the Editor can show a visual preview alongside the text. See Project Overview for details on that feature. The PDF this article covers is your actual translatable content, not a preview aid, and it gets automatically converted rather than uploaded as reference material.

Text-based PDFs (ones exported from a word processor or design tool, with selectable text) convert reliably in the vast majority of cases. Scanned PDFs — photographed or scanned paper documents with no underlying text layer — depend on OCR to recognize the text at all, so results vary more with scan quality, font, layout complexity, and language. A PDF with a large number of tables can also be harder to convert cleanly than plain text-heavy pages.
​

As practical guidelines rather than hard platform rules, keep uploads within roughly 100 MB total and 400 pages per document (around 150 pages for a scanned document) — very large or densely-tabled files are more likely to hit the underlying conversion service's own limits. If a file is large or complex enough to be at risk, splitting it into smaller sections before upload avoids the problem entirely.
​

If conversion fails outright, wxrks removes the file from the upload list and shows: "Conversion error: the file is too heavy or has a very complex layout. Please split the file or upload the .docx version." If you hit this, either split the PDF into smaller files and re-upload each one, or convert it to DOCX yourself (for example, in Word or Adobe Acrobat) and upload that instead — a manually-converted DOCX still goes through wxrks' normal DOCX parsing, just without the automatic PDF step.

How to create a translation project in wxrks — the full project creation walkthrough this article's Upload Files step is part of.
​Project Overview — covers every project tab, including the Reference Files tab's other upload types.
​How to work with a multilingual .xlsx file in wxrks — a different file-handling flow, for spreadsheets with one column per language.