How to find and merge duplicate contacts in a CSV file
CSV is what almost every tool exports: CRMs, newsletter platforms, event signup forms, phone backups. It's also where duplicates pile up, because the same person gets added from three different places with slightly different details.
Finding the duplicates is half the job. The other half is not merging people who just happen to look alike, like two colleagues with the same office phone or two different David Millers.
Step by step
- Check the header row
Make sure the first row has column names such as Name, Email, Phone and Company. Separate First Name and Last Name columns are fine too.
- Decide what identifies a person
A shared personal email or mobile number is strong evidence. A matching name alone is weak, because common names repeat. Company helps, but colleagues share it.
- Open the file in Contact Dedup
Drop the CSV onto the page. It detects the columns from the headers (you can correct them), then groups records that belong to the same person and explains each match.
- Review the uncertain pairs
Pairs without a shared email or phone are never ticked automatically. Have a look at them and tick the ones you agree with.
- Download the cleaned CSV
Each group becomes one row. The original file isn't changed. The last column lists which lines were merged, so you can check the result.
Things to watch out for
- If the CSV looks garbled when opened (strange characters instead of accented letters), it's an encoding issue. Contact Dedup reads both UTF-8 and the Windows encoding Excel often uses.
- European versions of Excel save CSV files with semicolons instead of commas. That's fine, the separator is detected automatically.
- Shared mailboxes like info@ or sales@ say nothing about who a person is. They are ignored when matching.
Questions
How big can the CSV be?
Without an account, up to 500 contacts; with a free account, up to 5,000. Larger files will come with the Pro plan.
What if my CSV has extra columns?
They are kept. When rows are merged, empty cells are filled from the other rows in the group.
Can it tell two different people with the same name apart?
Yes, as long as their emails or phones differ. Same-name pairs without other evidence go to review instead of being merged.
More guides
- What Contact Dedup understands, with examplesWhich pairs are the same person, and which are kept apart.
- How to remove duplicate contacts in Excel, including the ones Excel missesRemove Duplicates only catches identical rows. How to find the rest.
- How to clean up duplicate contacts from your phone (vCard / .vcf)Export, merge without losing photos or notes, import back.
- How to dedupe a Google Contacts exportWhen Google's Merge & fix isn't enough.
- Cleaning duplicate contacts out of a CRM exportCustomer data, sales history and why you shouldn't upload it.
- How to remove duplicates from a mailing list before you import itFewer double sends, cleaner stats, lower bills.
- How to find duplicate companies in a CRM or spreadsheetLegal forms, websites and branches in different cities.