Redacting the visible content of a file — blacking out a face, a street sign, a sensitive paragraph — says nothing about the metadata stored alongside it. A photo's EXIF block can carry the exact GPS fix, device model and capture timestamp; a document's properties and revision history can carry the real author, an internal username, and text that was "deleted" from the page but never purged from the file.
Stripping that metadata (scrubbing EXIF, resetting document properties, flattening revision history) is a separate step from redacting pixels or text — skip it, and a recipient who thinks to inspect the file learns everything the redaction was meant to hide.
- Redaction — hides sensitive content a viewer can see directly.
- Metadata scrubbing — removes sensitive content hiding in the file's hidden fields.
- Both are required; either alone leaves a leak.