De-duplication is the work of identifying copies and near-copies of the same asset and consolidating them. It is not the same as version control: versioning intentionally keeps a history of one asset, while de-duplication removes unintended copies that clutter the library.
Why it matters
Duplicates make search worse and storage bigger, and they create confusion about which file is authoritative. Clearing them is essential to a real single source of truth, and it is a standard step before any migration.
How it shows up in practice
When gathering scattered assets into one place before a migration, a team culls exact duplicates and obvious near-duplicates so only the right files move into the new DAM. Some platforms flag likely duplicates at ingestion to stop the problem recurring.
Common mistakes
- Migrating everything first and planning to clean up later, which rarely happens.
- Deleting a duplicate that was actually a needed version.
- Treating de-duplication as one-time rather than an ongoing ingestion guardrail.
Stacks covers cleanup in how to clean your dirty data quickly.