JSON to CSV Converter: Convert JSON to CSV Online | Free Tool

A Yes, column headers follow your JSON keys exactly as they appear in the first object. If your first entry has "name" before "email", that order sticks. But there's a catch: different browsers may reorder keys when displaying the preview. Download the file and open it in a text editor first to verify. For predictable results, use a JSON array with consistent key ordering across all objects.

A Not really. Browsers cap memory at around 2GB, so arrays over 100,000 rows will likely freeze your tab. For big datasets, split the JSON into 50,000-row chunks or use a command-line tool like jq. Test with 100 rows first—that takes under a second. Your browser's performance also depends on how much RAM your machine has.

A Yes, it preserves UTF-8 characters exactly. Accented letters, Chinese characters, and even emoji like 🚀 stay as-is in the CSV output. One caveat: some spreadsheet apps might mangle these on import. If your CSV shows garbled text, open it in a plain text editor first. For Excel users on Windows, save the file with UTF-8 BOM encoding—our download menu has that option.

A Yes, but you'll end up with lots of empty cells. The tool scans the first 50 objects to build its column list. If object 51 introduces a brand new key, that field gets dropped silently. For a complete CSV, either standardize your JSON keys first or run a quick script to find all unique keys across your entire dataset.

A Yes, but the tool flattens nested structures using dot notation. For instance, {"user":{"name":"Sam"}} becomes a column named user.name with the value Sam. Arrays inside objects get serialized as JSON strings in a single cell. That's usually fine for spreadsheets, but it won't give you relational tables. If you need true nesting, flatten your JSON manually first. This approach keeps the output predictable and avoids surprise empty columns.

A Yes, but you'll get a one-row CSV. The tool treats a single object like a one-element array, so all its key-value pairs become columns. For example, {"id":1,"name":"Sam"} converts to a CSV with headers id and name, and just one data row. That's rarely what people want—most use cases need multiple rows. If you're building an API response, wrap your object in square brackets before pasting it in. You'll save yourself a confusing pivot later.

A Paste the raw JSON response into the input box, then hit Convert. The tool auto-detects arrays, so a typical API payload like [{...},{...}] turns into rows instantly. If your JSON has a top-level wrapper like {"data":[...]}, strip that wrapper first—just copy everything inside the square brackets. Hit Download and you'll get a .csv file ready for Excel or Google Sheets. For API responses with deeply nested fields, expect dot-notation headers like user.address.city. That's fine for flat tabular use, but you won't get relational tables out of it. Start with a small sample payload to check the column order before converting the full dataset.

A No, that's a myth worth busting. When you paste JSON like [{"id":1,"name":"Sam"}] into the tool, the output keeps id and name as separate columns. Copy from the preview and commas stay intact. One gotcha: pasting straight into Google Sheets sometimes merges everything into column A. Use the Download button instead, then import the .csv file. That route preserves all column breaks exactly.

A Excel guesses data types automatically on import. If your JSON values look like numbers, opening the downloaded .csv file directly can strip leading zeros or turn long IDs into scientific notation. Try opening a blank spreadsheet first, then use the Data tab to import text columns manually. That forces Excel to respect your original formats.

A Not if you tell your software what delimiter you used. If you set a semicolon instead of a comma, Excel might dump everything into column A until you fix it. Go to Data, select Text to Columns, and pick your custom character. That fixes the layout instantly. Most spreadsheet tools handle custom characters fine once you configure the import wizard correctly.