Duplicate contacts come from three places: imports, web forms, and people adding a contact by hand without searching first. Merging the ones you have can be a one-afternoon job. The part that matters is closing each of those three doors, or the list fills up again within a few months and nobody trusts the contact history.
Why duplicates are worth fixing
Two records for one client means two half-histories. One colleague sees the quote, another sees the complaint, and the client gets a follow-up email about a job they already canceled. Reports count them twice. And when you finally merge them, someone has to decide which phone number is current.
Where they come from
Imports. The same person appears twice in the spreadsheet, or the sheet is imported a second time after a correction. If the CRM matches on fields that are blank or different in the sheet, every row becomes a new contact.
Web forms. A returning customer fills in your contact form with a different email address, a nickname or a typo. The form tool has no way to know it is the same person and creates a new record. Some form tools also create a new contact on every submission unless you set them to update existing ones.
Manual entry. Someone takes a call, types the name in, and saves. They did not search first because it was quicker to type.
Shared details. Two different people at one company entered with the same office phone or a shared address like info@ can look like duplicates when they are not. These need the opposite treatment: keep them apart and give each person their own details.
Find them
Export your contacts to a spreadsheet, even if the CRM has its own duplicate finder. Sorting by hand shows you the cases a tool cannot judge.
Sort by each of these in turn and look for neighbors that match:
| Sort by | What a match means |
|---|---|
| Very likely the same person. Check for shared addresses (info@, office@) first | |
| Phone (digits only, no spaces, brackets or country code) | Likely the same person, unless it is a switchboard |
| Company + last name | Worth a look: could be the same person with a new email |
| Name only | Weak. Two John Smiths are often two people |
Clean up phone numbers into one format before sorting. "+1 555 010 0100" and "(555) 010-0100" will not sort next to each other, and in a list built from phone calls, phone is the field to check most carefully.
Mark each pair as merge, keep separate, or not sure. Ask whoever owns the client relationship about the "not sure" ones.
Merge without losing history
Before merging anything, export the full contact list and keep the file. In many CRMs a merge cannot be undone.
For each pair:
- Pick the primary record: the one with the most history (deals, notes, emails), not necessarily the newest.
- Check which details are current. The newer record often has the current phone or email. Decide field by field.
- Merge, then open the result and check that deals, notes and activities from both records are there.
Companies need one extra check. Two records with the same company name can be two branches or two offices with different contacts and different billing. Merge them only when the address matches as well, or when you know they are one business with one point of contact.
Work through the list in one or two sessions. A half-merged database is harder to work with than either the clean or the messy version.
Stop new duplicates
None of these rules needs software:
- Search before you add. Search by phone or email, not by name, before creating any contact. Make it the rule for everyone who adds records.
- One person owns imports. Imports happen once per file, by one person, after a test with ten rows. People and companies go in before deals, and deals go in once.
- Forms update, not create. If your form tool or integration can match on email and update the existing contact, switch that on. Check what it does when the email is new but the phone matches.
- One phone format. Agree on one format and fix it on entry.
- No shared addresses as personal email. info@ goes on the company record, not on a person.
- A monthly five-minute check. Sort by email and phone once a month and merge what you find. Small batches stay small.
If the duplicates came from moving a spreadsheet into a CRM, the cleanup before the import matters more than any of the above; the same rules apply, just earlier.
In Pipedrive
Pipedrive has a Merge duplicates tool that lists likely duplicate people and organizations. According to Pipedrive's description of how it identifies duplicates, it flags people with the same name plus the same phone, the same email or the same organization, and organizations with the same name plus the same address. That means a "Jon Smith" and a "Jonathan Smith" with the same email will not show up in the list; your email sort will catch them.
Deals, people and organizations can also be merged by hand from the item itself. Only admins or users with the merge permission can do it, the primary record wins where values conflict, and Pipedrive says items merged through the duplicates tool cannot be un-merged. Export first.