Remove metadata from documents
Most tools in this category handle photos or Office files, not both. This one handles PDF, Word, Excel and PowerPoint under the same engine and the same no-upload guarantee as every image format — because a document you're about to send externally deserves the same scrutiny as a photo.
Which format leaks what
PDF and the three Office formats share more structure than you'd expect — Word, Excel and PowerPoint are all built on the same packaging convention, so the same categories of risk (properties, comments, edit history) show up in all three, just under different names.
| Format | Can carry | Worst case |
|---|---|---|
| The Info dictionary (Author, Title, Creator) and a separate XMP packet | Author name readable in the XMP packet even after clearing Document Properties | |
| Word (.docx) | Core/extended/custom properties, comments, tracked-change authorship | A sentence someone deleted, still in the file, attributed and timestamped |
| Excel (.xlsx) | Core/extended/custom properties, cell comment threads | Company name and editing time from every person who touched the sheet |
| PowerPoint (.pptx) | Core/extended/custom properties, comment threads, the source template name | An internal template name that outs a deck as not built from scratch |
The one that ends careers
Document properties are the least of it. If track changes was ever turned on in a Word file, deleted text often survives inside word/document.xml — attributed by name, timestamped, and invisible unless you know to look for it. Clearing "Document Properties" doesn't touch it; only a tool that reads the tracked-change markup itself does. The same risk exists in comment threads across all three Office formats: a thread resolved and hidden in the UI is still sitting in the file underneath.
Pick your format
- Remove the author from a PDF — clears the Info dictionary and the separate XMP metadata stream most other tools miss.
- Remove metadata from a Word document — tracked-change authorship and deleted text, comments, and every properties block.
- Remove metadata from an Excel spreadsheet — properties and cell comments, with formulas and formatting untouched.
- Remove metadata from a PowerPoint deck — properties and comment threads, with slides and speaker notes preserved.
Why Word, Excel and PowerPoint share the same risk shape
A .docx, .xlsx or .pptx file is a ZIP archive of XML parts — rename any of them to .zip and you can open it directly. All three follow the same packaging convention, the Open Packaging Conventions defined in ECMA-376, the Office Open XML standard. That shared foundation is why the fix looks almost identical across all three formats: docProps/core.xml and app.xml hold the same creator, company and editing-time fields regardless of whether the file is a document, a spreadsheet or a deck, and each has its own equivalent of a comment thread wired into the same properties schema.
PDF doesn't share that packaging, but it shares the underlying shape of the risk: a structured properties block most tools clear, and a second, separate metadata mechanism — the XMP packet — that duplicates the same fields and that most tools don't check at all.
One engine, one guarantee
Every format above is cleaned by the same sandboxed frame that handles photos, served with the same Content-Security-Policy of connect-src 'none'. The one difference: removing PDF and Office properties means re-serializing the file, so unlike photos we can't offer a byte-identical hash — and we say so on the result rather than implying otherwise.
Cleaning a photo instead? See the image metadata hub for JPG, PNG, WebP, GIF, TIFF and BMP, or the full ten-format comparison for both families side by side. Sending a whole folder of mixed files? The batch cleaner handles documents and photos together, with no file limit.