How to find duplicate email addresses in a CSV without deleting the wrong rows
A review-first workflow for locating repeated email addresses, distinguishing true duplicate contacts from legitimate history, and exporting a safe cleaned copy.
Before you work through this by hand: you can check your own CSV free and see which of these problems your file actually has — duplicate rows, near-duplicates, broken quoting, encoding damage. It runs in your browser tab, nothing is uploaded, and there's no signup or email.
Start with a copy and define what the email means
Keep the original export unchanged and work from a dated copy. A repeated email address can mean two duplicate contact rows, or it can be legitimate history such as multiple orders, subscriptions, or event registrations. Decide whether you are deduplicating contacts, transactions, or another record type before removing anything.
Find exact repeats in a spreadsheet
In Excel or Google Sheets, select the email column and use conditional formatting to highlight duplicate values. Then filter the highlighted rows and inspect the surrounding fields: name, customer ID, status, last-updated date, and source system. Highlighting is a review aid; it is not evidence that every repeated address should be merged.
Check duplicates with Python
import pandas as pd
df = pd.read_csv("contacts.csv")
email = df["email"].astype("string").str.strip().str.casefold()
repeated = df[email.duplicated(keep=False)].assign(_email=email)
print(repeated.sort_values("_email"))
Normalize only the matching key. Keep the original email text visible for audit, and treat blank emails separately rather than grouping every blank as one person.
Separate duplicate contacts from legitimate repeated records
Two rows with the same email and the same contact identity are strong duplicate candidates. Two rows with the same email but different order IDs, timestamps, or transaction amounts may be valid separate records. If fields conflict, flag the group for review and preserve the losing values instead of keeping the first row automatically.
Review near-duplicates too
Exact email matching misses whitespace, capitalization, and occasionally export-specific variations. It also cannot tell whether aliases such as [email protected] and [email protected] belong to the same person. Treat those as suggestions that need context, not as automatic deletions.
Export a new file and verify the result
Record the source row count, the groups reviewed, and the rows intentionally removed or merged. Export to a new filename, reopen it with a CSV parser, and verify the header, row count, and a sample of retained records. CSV Repair & Deduper runs this review-first process locally in your browser: it flags exact and near-duplicate candidates, keeps the original values visible, and lets you approve, edit, undo, and export without uploading the file.
Try the free check on your own CSV first. See also our comparison of deduplication and manual review and guide to conflicting duplicate data.
See CSV Repair & Deduper (offline browser tool) — $19 one-time