Duplicate Row Finder

Find duplicate rows in a CSV or table

Paste a CSV (or any delimited table) and this tool finds the rows that repeat, groups them together with their line numbers, and tells you which columns are driving the duplication. It is read-only: it reports the duplicates, it does not delete or rewrite anything. Everything runs in your browser — nothing is uploaded.

By default two rows match only when the whole row is identical. Give one or more key columns to match on just those (e.g. treat rows with the same email as duplicates even when the name differs). Turn on Ignore letter case and Ignore extra whitespace to catch near-duplicates that differ only by casing or stray spaces.

Worked example

Input (with the header row on):

name,email
Alice,[email protected]
Bob,[email protected]
Alice,[email protected]
Carol,[email protected]
Bob,[email protected]

Report output:

Scanned 5 data rows (2 columns), keyed on the whole row.
Found 2 duplicate groups covering 4 rows; 1 row is unique.

Duplicate groups (most repeated first):
  x2  Alice | [email protected]   (lines 2, 4)
  x2  Bob | [email protected]   (lines 3, 6)

Columns driving duplication (repeated rows / distinct values):
  name: 4 / 3
  email: 4 / 3

Line numbers count from the top of the input (the header is line 1), so you can jump straight to each duplicate in your source file.

Options

• Key columns — header names (when there's a header) or 1-based indices, comma-separated. Blank matches the entire row.
• First row is a header — use it to name columns and exclude it from the scan.
• Delimiter — comma, tab, semicolon, pipe, or any single character.
• Ignore letter case / extra whitespace — normalize before comparing so near-duplicates that differ only by case or spacing still group together.
• Output — Report (a readable summary plus the column analysis), Duplicate rows only (CSV) (just the offending rows, ready to paste back), or Structured (JSON) (groups + per-column stats for scripting).

Limits & edge cases

• Comparisons are exact after the case/whitespace normalization you choose — this catches formatting differences, not typos. For typo-level fuzzy matching use the fuzzy-dedupe tool instead.
• Blank cells are ignored in the "columns driving duplication" analysis (an empty value is not a meaningful duplicate), but they still count as part of a whole-row or key match.
• Rows with fewer fields than expected treat the missing cells as empty.
• Everything is processed in memory in your browser; very large files are bounded by your device's memory.

FAQ