JSON to CSV

An array of flat objects with consistent keys: `[{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]`. The first object's keys become the CSV column headers, and every subsequent object fills one row. Objects don't all need to have every key — missing values become empty cells. If your data is wrapped in a parent object like `{"users": [...]}`, drill down to the array first (e.g., `data.users`) before pasting. For deeply nested or hierarchical JSON, you'll typically want to flatten it first since CSV is a flat tabular format.

Nested values are serialized as their JSON string representation in the cell. So `{"name": "Alice", "address": {"city": "NYC"}}` produces a column called `address` with the literal value `{"city":"NYC"}`. This preserves the data but isn't usually what you want for analysis. The cleaner pattern: flatten the structure first (e.g., make `address.city` its own column) using a tool like jq, or transform the JSON in your code before pasting. CSV is fundamentally flat, and forcing nested data into it always involves trade-offs.

No — the converter follows RFC 4180 quoting rules. Any value containing a comma, double-quote, or newline gets wrapped in double quotes, and embedded double-quotes inside that value are escaped by doubling them (`"` becomes `""`). So a value like `She said "hi, world"` becomes `"She said ""hi, world"""` in the CSV, which Excel, Google Sheets, and any standards-compliant parser will read back correctly. If you're seeing fields split across columns when you open in Excel, the original tool that generated the CSV probably wasn't quoting properly — this one is.

That's an Excel quirk: by default, Excel on Windows opens CSV files using the system's legacy encoding (often Windows-1252), not UTF-8, so any non-ASCII characters look like mojibake (`å¼µ` instead of `張`). The fix is to either (a) open Excel first and use Data → From Text/CSV → choose UTF-8, or (b) prepend a UTF-8 byte-order mark (BOM, `\uFEFF`) to the file, which signals UTF-8 to Excel. This converter outputs plain UTF-8 without a BOM by default to keep the file portable. Google Sheets and Numbers handle UTF-8 correctly without any workaround, so if you have the choice, open the file there.

Yes, up to several megabytes — the conversion runs in browser memory using JavaScript's native JSON parser plus a streaming CSV writer, which handles 50,000+ rows on a modern laptop without breaking a sweat. For truly massive exports (multi-million rows, hundreds of MB), you'd want a streaming command-line tool like `jq -r` piped to a CSV writer, since the browser's memory headroom isn't designed for that scale. But for everything in the API-response or report-export range — typical Stripe, Shopify, or Postgres exports — this handles it fine and keeps the data on your device.