JSON to CSV Converter

How This Tool Converts JSON to CSV

• Accepts a JSON array of objects or a single object; each array element becomes one CSV row.
• Column headers are collected from all unique keys across every object — if objects have different keys, missing cells are left empty.
• Nested objects are flattened automatically using dot notation: address.city and address.country each become their own column.
• Array values inside objects (e.g. "roles": ["admin","editor"]) are joined with | into a single cell value.
• Any value containing a comma, double quote, or newline is wrapped in double quotes (and internal quotes are doubled), following the RFC 4180 CSV rules that Excel and Google Sheets expect.

When to Use JSON to CSV

• When you need to open API response data in Excel or Google Sheets for review or sharing.
• When preparing data for bulk import into a database, CRM, or data warehouse that expects CSV format.
• When building a report and your data source is a JSON API response array.
• When handing structured data to a non-technical stakeholder who prefers a spreadsheet view.

Worked example

Here is a realistic array that exposes the three things people most often get wrong — nested objects, arrays inside a record, and objects that don't all share the same keys:

Input JSON

[
  {
    "id": 1,
    "name": "Alice",
    "address": { "city": "Boston", "zip": "02101" },
    "roles": ["admin", "editor"]
  },
  {
    "id": 2,
    "name": "Bob, Jr.",
    "address": { "city": "Denver" }
  }
]

Output CSV

id,name,address.city,address.zip,roles
1,Alice,Boston,02101,admin | editor
2,"Bob, Jr.",Denver,,

What happened, line by line

• The nested address object was flattened into two columns, address.city and address.zip.
• The roles array was collapsed into one cell as admin | editor — a pipe-joined string, not two columns.
• The headers are the union of both objects' keys. The second record has no address.zip and no roles, so those two trailing cells are left empty (note the two commas at the end of the last row).
• "Bob, Jr." contains a comma, so it was wrapped in double quotes to keep it in a single column.
• 02101 is preserved exactly as written in the CSV — but see the leading-zero gotcha below, because Excel may still mangle it on open.

Edge cases & gotchas

• Excel drops leading zeros. A ZIP code or ID like 02101 is correct in the CSV file, but Excel auto-detects it as a number and shows 2101. To keep it, import via Data → From Text/CSV and set that column's type to Text rather than double-clicking the file.
• Excel and UTF-8. The download is UTF-8, but Excel on Windows assumes the legacy ANSI codepage on a double-click, so accents and emoji show as garbled text. Tick UTF-8 BOM (Excel) before downloading, or import with Data → From Text/CSV and choose UTF-8 as the file origin.
• Arrays of objects become raw JSON. Primitive arrays join cleanly (admin | editor), but an array of objects is serialized back to JSON strings inside the cell. Flatten or extract those before converting if you need real columns.
• Deeply inconsistent objects produce sparse rows. Because headers are the union of every key seen, a set of very different objects yields a wide table with many empty cells — expected, not a bug.
• Order follows first appearance. Columns appear in the order keys are first encountered across the array, so the first object effectively sets the leading column order.