Drop a file, choose which columns make a row unique, and download the cleaned copy. The file is read by your browser and never sent anywhere.
All columns selected means only fully identical rows count as duplicates. Select just an ID or an e-mail to catch rows that repeat a key but differ elsewhere.
Why this page has no upload button. De-duplicating rows is string comparison — there is no reason it needs a server. Most online CSV cleaners upload first anyway, which rules them out for customer lists, payroll or anything under an NDA. This page has nowhere to upload to: open your network tab and drop a file, the request list stays empty.
By default every column is part of the key, so a row is only removed when it repeats another row exactly. That is the safe setting. In practice, duplicates usually differ in some trailing field — the same customer imported twice with a different note, or the same order with a later timestamp. Selecting only the columns that identify the record (an order number, an e-mail address) catches those.
Keep the first preserves the earliest copy in file order; keep the last preserves the most recent, which is usually what you want when later rows carry updated values.
| Delimiter | Comma and semicolon, detected automatically. The cleaned file is written back with the same delimiter it came in with. |
|---|---|
| Quotes | Quoted fields containing commas and line breaks, with "" escaping. Re-quoted correctly on the way out. |
| Line endings | LF and CRLF are read; a leading UTF-8 BOM is stripped and written back so Excel opens the file correctly. |
| Comparison | “Ignore case & extra spaces” trims each cell, collapses runs of whitespace and compares case-insensitively — so Ada Lovelace and ada lovelace match. |
| Size | Held in memory. Comfortable up to a few tens of thousands of rows. |
Does the file leave my computer? No. There is no server behind this page. The file is read with the browser's File API and the comparison happens in the page. Once loaded, it keeps working with the network off.
Which copy is kept? “Keep the first” preserves the earliest copy in file order, “keep the last” the most recent — usually what you want when later rows carry updated values.
Will Excel open the cleaned file correctly? Yes. The delimiter, the line endings and the byte order mark are all written back exactly as they came in, so accented and Turkish characters survive the round trip.
Can it read .xlsx? Not on this page — it reads CSV only. For Excel workbooks use Sheet Insights.
Sheet Insights reads the same file and reports what is wrong with it: missing values by column, type mismatches, outliers, leading-zero codes that Excel has eaten — then cleans them with reusable recipes. Same rule — everything runs in your browser, with no server behind it.
See what Sheet Insights does