Account

What new tool do you want?

Developer Tools

JSON to CSV Converter

Turn a JSON array of objects into CSV for Excel or Google Sheets — nested objects are flattened into dot-notation columns.

How to use

  1. 1Paste a JSON array of objects.
  2. 2Copy or download the CSV and open it in any spreadsheet.

Frequently asked questions

How are nested objects handled?

They're flattened into columns using dot notation, so { "user": { "name": "A" } } becomes a user.name column.

What if objects have different keys?

Every key found in any object becomes a column; missing values are left empty.

How are arrays inside objects exported?

Arrays are written as JSON text in a single cell, so no data is lost.

The column set is the union of every object's keys, not just the first

Rather than taking columns from the first object in the array (a common shortcut that silently drops data when later objects have extra fields), this tool collects every key seen across every object in the array before building the header row, then fills in a blank cell for any object missing a given key. That's why the example input (one object with a `user.name` field, another with a `tags` field) produces a header row containing both `user.name` and `tags`, with each row's missing column left empty rather than throwing an error or dropping the field.

Flattening is dot-notation, recursively, with one escape hatch for arrays

Nested objects are flattened by recursively joining keys with dots (`{ "user": { "name": "A" } }` → a `user.name` column), which works for arbitrarily deep nesting. Arrays are handled differently on purpose: rather than trying to flatten `tags: ["a","b"]` into `tags.0`, `tags.1` columns (which breaks down when rows have different array lengths), the whole array is serialized as a JSON string in one cell — `["x"]` — so the data is fully preserved and re-parseable, at the cost of not being natively sortable/filterable as separate spreadsheet columns.

Why there's a semicolon delimiter option

The semicolon delimiter exists specifically for European-locale Excel, which uses the comma as the decimal separator and therefore expects semicolons to separate CSV fields — opening a comma-delimited file in that Excel configuration mangles numbers and misaligns columns on import. If your CSV is headed for a European spreadsheet rather than a database or script, switching the delimiter before downloading avoids that import step entirely.

Worked example: JSON to CSV Converter

Input `[{"id":1,"user":{"name":"Ann, B"}},{"id":2,"tags":["x"]}]` produces three columns — `id`, `user.name`, `tags` — and this CSV: id,user.name,tags 1,"Ann, B", 2,,"[""x""]" Row 1 has no `tags` value so that cell is empty; row 2 has no `user.name` so that cell is empty; the comma inside 'Ann, B' is safely quoted, and the array `["x"]` is written as a quoted, escaped JSON string in a single cell.

A lone object becomes one row, and an array stays one cell

If I paste a single object instead of a list, it is treated as one row. A value that is not an object at all lands in a column named value. Nested objects become dotted column names. Arrays are not split into columns. They are written as one cell, and any quote inside that cell is doubled and the cell is wrapped in quotes. A key that is missing on a later row is an empty cell, not the word null. The delimiter opens on a comma. Semicolon is labeled for European Excel, and tab is the third choice. A comma, a quote, or a line break inside a cell is quoted. Invalid JSON stops with a line and a column and produces no table.

Related tools

More: Developer Tools