CSV Cleaner
Auto-detect delimiter, encoding, and column types, then clean and export your CSV
All processing happens in your browser. Your data stays private.
Upload CSV File
Drop files here or click to browse
Accepts: .csv
Max size: 10MB
How to Use
- 1Upload your CSV file (up to 10MB)
- 2Review the detected delimiter, encoding, and any malformed rows found
- 3Check the detected column types and remove invalid rows per column if needed
- 4Select cleaning options (remove duplicates, trim whitespace, etc.)
- 5Choose which columns to keep
- 6Click 'Clean CSV' to process your data
- 7Download as CSV, JSON, Excel, or a Markdown table
Tips
- •If the preview looks garbled, use the encoding retry button instead of re-uploading
- •Column type detection is a heuristic; review flagged rows before removing them
- •Removing duplicates compares entire rows for exact matches
- •Empty columns are only removed if every cell in that column is empty
- •Special character removal keeps alphanumeric characters and spaces
- •Markdown export is handy for pasting a data table into documentation or a README
- •JSON export is useful for importing into databases or APIs
Common Use Cases
Data Import Cleanup
Remove empty rows and trim whitespace before importing into a databaseExcel Export Fix
Clean Excel exports that contain extra empty rows and columns at the endDeduplication
Find and remove duplicate entries from customer lists or inventory dataData Validation
Catch malformed emails, dates, or phone numbers before importing a mailing listReport Conversion
Turn a raw CSV export into a Markdown table or Excel file for a reportFrequently Asked Questions
What does the CSV cleaner do?
It detects delimiter and encoding, infers column types (email, URL, date, phone, number, boolean), removes empty rows and columns, trims whitespace, drops duplicates, and exports to CSV, JSON, Excel, or Markdown. Use it to clean messy CSVs before analysis or import.
How does delimiter and encoding detection work?
The delimiter (comma, semicolon, tab, or pipe) is auto-detected from the file's structure; if it guesses wrong, pick one manually from the dropdown. Encoding is detected by checking for a byte-order mark and, failing that, decoding as UTF-8 and checking for replacement characters, a sign the file is actually Windows-1252 (common from older Excel exports). If text still looks garbled, use the "Try [other encoding]" button to re-decode without re-uploading.
How accurate is the column type detection?
It's a best-effort heuristic, not a validator: a column is labeled with a type only when at least 80% of its non-empty values match that pattern. Review the flagged invalid rows before removing them, especially for date and phone columns, which use permissive matching and can have false positives on free text.
What export formats are available?
CSV, JSON, Excel (.xlsx), and a GitHub-flavored Markdown table. Excel and Markdown are useful for reports and documentation; JSON for importing into an application or API.
Will my CSV file be uploaded to a server?
No. All CSV processing, including parsing, type detection, and export, happens entirely in your browser using JavaScript. Your data never leaves your device.
What cleaning options are available?
You can skip malformed rows, remove rows with invalid values in a chosen column, drop empty rows and columns, trim whitespace, remove duplicates, and normalize case. Toggle each option as you need.
Is there a file size limit?
The tool can handle most CSV files, but very large files (over 50MB) may take longer to process depending on your device. All processing uses your browser's memory.