Find and merge duplicates

Review likely duplicate people, households, and companies side by side, and merge each pair in one confirmed click.

2 min read

Duplicates creep in through imports, forms, and honest retyping. They split a person’s history across two half-records. A nightly sweep hunts them down across people, households, and companies (imports catch most on the way in; this queue is for what slips through), and the Duplicates page is where you review what it found.

Review and merge

Open DuplicatesChoose people, households, or companies.
Read the confidence and the why-flagged reasonEach pair is labeled High confidence or Possible match, with a sentence naming what matched (same email, same name at the same address, and so on) and a side-by-side comparison of the fields that differ.
Merge into one, or Not duplicatesMerging fills blanks on the record you keep from the one you remove (it never overwrites a value that is already there), and you confirm before anything happens. Genuinely two different people? Choose Not duplicates and the sweep will not flag that pair again.

Merges are permanent

The duplicate record is removed for good. The confirmation names both records so you know exactly what is merging into what. When unsure, open both profiles first.

Caught a pair in a grid instead? Select exactly two rows and use Merge in the bulk action bar. Same result, no trip to the finder. See Selection, bulk actions, and merging.

Merging two volunteers keeps only one volunteer record

A person can hold one companion volunteer record, so merging two people who both have one keeps the record you are keeping and removes the other, along with the phones signed in on it. If the removed one was the approved record, that volunteer verifies a code again and an admin approves them again before they can canvass or deliver. The merge confirmation tells you when this pair is affected. See Volunteer access approvals.

Make it a habit

A five-minute duplicates pass after every import keeps the database trustworthy, far cheaper than a heroic annual cleanup.

Try this on sample data.

Every feature in the docs is live in the free demo workspace — no card, nothing to lose.

Start free