CSV Duplicate Finder
You exported a customer list, a bank statement or a product feed, and something looks off: the same contact appears three times, an invoice was imported twice, the row count is higher than it should be. Opening the file in a spreadsheet and hunting by eye is slow and error-prone. Drop your CSV here, choose which columns define a duplicate, and see every repeated group at once. Then download a clean copy. Your file never leaves your device.
Your file stays in your browser. Nothing is uploaded anywhere.
No file selected. Up to 50,000 data rows are read.
Duplicate groups
Nothing to show yet.
Note: this is a planning aid. Results are estimates of what matches your settings and do not replace review by an accountant or tax professional if the data feeds your books.
Clean Data in Minutes – CSV & Excel Cleaning Toolkit (Python)
FAQ
Is my file uploaded anywhere?
No. The file is read locally by your browser and processed on your own device. There are no network requests, and the cleaned file is created in your browser when you click download.
How does it decide that two rows are duplicates?
Rows are duplicates when all the key columns you ticked hold the same value. Tick every column to find fully identical rows, or only one column such as an email address or invoice number to find repeated records that differ elsewhere.
Which row is kept in the cleaned file?
The first occurrence in the file is kept and later matches are dropped. The original order of the remaining rows is preserved. If the later row is the better one, sort your file first and run it again.
Does it handle semicolons, tabs and quoted fields?
Yes. Comma, semicolon and tab delimiters are detected automatically from the start of the file, and you can override the choice. Quoted values containing the delimiter, doubled quotes or line breaks are read correctly and written back with proper quoting.
Why are ' Smith' and 'smith' treated as the same?
By default the tool ignores spaces around values and upper/lower case, since these are common causes of hidden duplicates. Untick those options if differences in spacing or capitalisation should count as distinct values.
Is there a size limit?
The tool reads up to 50,000 data rows to keep your browser responsive. If your file is longer, a message tells you, and the download contains only the rows that were read. Split very large files into parts first.
Runs entirely in your browser – nothing is sent or stored.