How to Find Duplicate Files in OneDrive

Start with inexpensive signals. Byte size quickly narrows a large inventory, but it cannot prove equality by itself. Filename is even weaker because sync conflicts, manual renames, and generic names create false matches. Use these fields to find candidates, then require a reliable fingerprint before calling two files exact copies.

OneDrive can expose hashes through its file metadata. For supported files, quickXorHash gives a content-derived value that can be compared without transferring the file body to the scanner. A size-and-hash match is strong evidence of byte-identical content; it still says nothing about which folder, permissions, or history matters most.

Open each exact group as a decision, not as a delete queue. Prefer a copy in the canonical project folder over one in Downloads or an abandoned import folder. Check whether any copy is shared, whether a workflow points to its path, and whether its version history contains information you still need.

Keeping multiple copies is valid when they serve separate permissions or retention needs. The safest plan records one or more keep choices and lists the remaining candidates. Protected formats and ambiguous metadata deserve manual review even when the fingerprint agrees.

A good cleanup plan names the retained file, the candidate copies, their paths, sizes, and the evidence used. Review the largest groups first because they produce the clearest storage benefit with fewer decisions. Save the plan before granting any write permission.

If you later execute a plan, re-check each item against the current drive state. Files can be moved, edited, or reshared between scan and cleanup. ClearCopies keeps the scan and review stage separate from any later write step so you can stop after the report if that is all you need.