Remove duplicate rows from CSV

For cleaning an extract that repeats the same record, Remove CSV duplicates keeps one row according to all or selected columns.

Deduplicate CSV rows across the whole table or only the columns that define a repeated record. You get the result directly in your browser.

How to use

Open the file you want to clean from the file picker, or drop its contents into the text area provided.
Check which separator the automatic read picked; set another one if the values arrive in the wrong fields.
Choose whether the check covers the whole row or only the columns that define one repeated record.
Run the removal: one line is kept for each combination and the later repeats leave the produced table.
Check what was dropped by comparing the source with the result, then export the cleaned table.

Example

A contact list merged from two platforms often holds the same person twice, with a stray space or a different capital letter. By limiting the check to the columns that truly identify a record, the produced table keeps one line per entry and sets the repeats aside.

Limits and formats

The comparison is literal, so two addresses written differently are treated as two separate records.
The first occurrence is the one that survives, and the page does not let you choose a different row to keep.
Input is a CSV or TXT file, and the original structure must stay consistent for the output to be usable.
Keep a copy of the source if the list is hard to rebuild, since the removal cannot be undone from the page.

Where processing happens

Cleaning happens inside your browser, so a contact list can be deduplicated without leaving your computer and without leaving a trace once the tab is closed.

Frequently asked questions

Are two nearly identical rows merged?

No. A single difference in spacing, capital letters or accents is enough to make the two rows count as separate records. Tidy the casing and the spare spaces in your spreadsheet first if you want them treated as one.

What happens when the check uses only one column?

Any row whose value matches a row already seen is discarded, even when the remaining columns differ. That is handy for keeping a single copy of an identifier, but it can throw away details you actually needed.

Does the row order change after removal?

The rows that survive keep the order they had in the source file. Only the repeats disappear, which makes it easier to compare the output with the original and confirm that nothing else moved.