Find duplicates

Last updated: 1 October 2026

A duplicate search compares the synced records of one type against your active match rules and puts every pair it finds in the review queue. It runs every night for every record type, and you can start it yourself at any time.

Start a search

  1. Sync first, so the search works on current data.
  2. On the Dashboard, find the record type in the Per-entity status table.
  3. Click the search button in the Duplicate detection column. A preview shows what the run will do and roughly how long it takes.
  4. Click Start detection. The run continues in the background.

While a run is going, its button changes to Stop. Stopping keeps the pairs already found. Only one run per record type can be live at a time.

Detection settings

Open Match rules and scroll to Detection settings.

  • Minimum fuzzy name score: a floor from 80 to 100 for every fuzzy name and organization name condition. Recommended: 92.
  • Ignored terms: one per line. Names equal to one of these terms, ignoring case, are left out of name matching. Useful for values like test or unknown.

Groups held back

When records share an email address or phone number that too many records in your account carry, and no name backs the match up, Surge Dedupe does not offer them as pairs. The Review page lists these groups with the shared value and the number of records, so you can see they are not missing by accident.

Start over after changing rules

  1. Open Match rules and go to Reset detected duplicates.
  2. Choose All entities or one record type.
  3. Click Clear pending duplicates and confirm with Clear pending.
  4. Run the search again from the Dashboard.

This deletes only the pairs waiting for review. Merged and skipped pairs are kept. It cannot be undone, but the pairs are found again by the next search.

Still stuck? Email info@sales-surge.nl