Remove Duplicates from CSV

Paste the CSV or open the file, tick the columns that identify a duplicate — an email or customer ID, say — or leave them all unticked to compare whole rows. Choose whether to keep the first or last copy, then download the cleaned file. Removed rows can be downloaded separately with the reason for each.

No. Every value is treated as text, so 00123 stays 00123, long numbers aren't rounded and dates aren't reformatted. Only the options you choose — trimming spaces, removing empty rows and duplicates — change anything.

Yes. It uses a full CSV parser, so a quoted field like "Smith, John" or a note spanning two lines stays one cell. If a quote is never closed you'll see a warning.

Yes — choose “Ignore case” and ANA@EXAMPLE.COM matches ana@example.com. The row you keep is written exactly as it was.

Comma, semicolon, tab and pipe. The delimiter is detected from the first lines, and you can override it. You can also choose the delimiter for the download.

No. The CSV is read and cleaned in your browser; nothing is sent anywhere or stored.

Keep first suits most lists, such as the original signup. Keep last is handy when newer rows were appended to the end of the file and you want the latest version of each record. Either way, the rows you keep stay in their original order — nothing is sorted.

Usually there's a hidden difference: a trailing space, different capitalization, or a space inside a value. Choose “Trim ends” or “Trim & collapse” and “Ignore case”, and compare only the columns that matter. Accented letters typed in different Unicode forms are already treated as the same.

It removes spaces at the start and end of every cell and turns any run of spaces, tabs or line breaks inside a cell into a single space. That also joins multi-line cells into one line, so use “Trim ends” if you need to keep line breaks.

Untick “First row is a header”. The first row is then treated as data and can be removed as a duplicate, columns are labeled Column 1, Column 2 and so on, and the removed-rows file gets generic column_1, column_2 headings.

They're kept as they are, not padded or cut, and the result panel tells you how many rows have a different number of columns from the header. Those rows can still be matched as duplicates.

You can open files up to 20 MB. The preview shows the first 200 kept or removed rows, and the downloads include all of them. Very large files take longer on older phones and laptops because everything runs in the browser.