Guide
Checking a document for hidden data before you share it
A document you send is more than the page you see. A Word file remembers who wrote it, how long they spent, who commented, which words were deleted with tracking on, and sometimes the company that licensed the software. A spreadsheet can hold hidden sheets, a presentation keeps speaker notes, and a PDF with black boxes may still contain the words underneath. This checker looks inside the file and reports what it finds, so you can clean it before it leaves your hands.
What is inspected
For Office and OpenDocument files, the page opens the package and reads the document properties (author, last editor, company, created and modified dates, revision number, total editing time), then walks through the content parts. It counts comments and who wrote them, tracked insertions and deletions with their authors, runs of hidden text, hidden sheets, slides and rows, speaker notes, external links, embedded objects, macros and the stored thumbnail. For a PDF it reads the Info dictionary and XMP packet, attachments, JavaScript, annotations and optional layers.
A worked example
The Sample button builds a small Word file called contract-draft.docx. The report shows three serious findings and two to look at. High: one comment, by Alice Smith. High: tracked changes with one insertion and one deletion by Bob Jones, with the warning that deleted text stays in the file until the changes are accepted or rejected. High: one run of hidden text. Medium: names stored in the file, Alice Smith and Bob Jones. Low: a company, Contoso Ltd, is recorded. The property table lists Alice Smith as author, Bob Jones as last editor, created 2026-01-05, modified 2026-10-30, revision 12, application Microsoft Office Word and 388 minutes of editing time.
Testing a redaction
Choose a PDF, then enter the words that should be gone, one per line. The page extracts the text layer of every page and reports each word it can still find. If a name appears, the black box is only drawn over it: anyone can copy it out with select-all or find it with a search. A proper redaction removes the content, not just the look. If the extractor finds no text at all, the page may be scanned images, and this test cannot say anything about it.
Cleaning up
Findings come with the place to fix them. In Word, Excel and PowerPoint, use File, Info, Check for Issues, Inspect Document. In PDF editors, remove metadata and flatten or properly redact. For the strongest result, copy the final content into a new file instead of editing an old one, or export to PDF and re-check that PDF too.
Limits
The checker cannot see metadata embedded in pictures, such as camera EXIF data, and it does not judge the visible text. A file that passes shows that none of the common leak sources was found, not that it is safe to publish.
Related tools
- Redact PDF removes text and images from a PDF page.
- PDF Metadata Editor changes or clears PDF properties.
- EXIF Data Remover cleans the metadata of photos.
- File Type Identifier shows what kind of file you actually have.
Checked against the current version of the Document Privacy Checker on .