CSV to JSON Converter
CSV rows become a JSON array, keyed by the header row.
Turn nested JSON into flat, dot-notation key/value pairs for CSV columns, spreadsheets or diffs.
Your browser is preparing the tool. It runs 100% locally.
The JSON Flattener collapses nested JSON into a single-level object where every key is a dot path, like user.address.city. Array items get a numeric index, so tags becomes tags.0, tags.1. Only the leaf values are kept; the structure becomes the key. It needs valid JSON and outputs valid, pretty-printed JSON, all in your browser.
It walks a nested JSON object and rewrites it as flat key/value pairs, where each key is the full path to a value. That is the shape you need when a target system expects columns, such as a CSV export, a spreadsheet, a BI tool or a SQL table.
It also gives you stable, predictable keys, which makes flattened JSON handy for searching logs by path or diffing two payloads to see what changed.
It parses your JSON, then recursively walks the tree. Each time it reaches a leaf value (a string, number, boolean or null) it records the full dotted path as the key.
Objects add their key to the path; arrays add the element's index. The result is serialised back to neatly indented JSON.
Nested object → parent.child (dot notation)
Array element → arrayname.index (for example tags.0)
Only leaf values are kept; the structure becomes the key{"user":{"name":"Alice"},"tags":["x","y"]} → {"user.name":"Alice","tags.0":"x","tags.1":"y"}Array indices use a dot (tags.0), not bracket notation (tags[0]). Keep that in mind if a downstream tool expects brackets.
Your browser parses and flattens the JSON. The site doesn't upload, store or log anything you paste.
There is no account or history; closing the tab clears your data, so it is safe for configs and API payloads.
| Delimiter | Dot (.) |
|---|---|
| Arrays | Numeric index (tags.0) |
| Keeps | Leaf values only |
| Input | Valid JSON |
| Reverse (unflatten)? | No |
| Processing | 100% in-browser |
It requires valid JSON. Comments, trailing commas, single quotes or unquoted keys cause a parse error rather than a guess.
Empty objects and empty arrays produce no key at all. They have no leaf values, so they disappear from the output.
Array indices are written with a dot (tags.0), not brackets (tags[0]). If a downstream tool expects bracket notation you will need to adjust.
If an original key already contains a dot, the flattened path becomes ambiguous, because a dot is also the separator.
It is one-way. There is no unflatten here to rebuild the nested structure from dotted keys.
Flatten an object so each dotted path becomes a column header for a CSV or Excel import.
Turn a deeply nested response into flat key/value pairs that map cleanly onto a table or database row.
Flatten both versions and compare the flat keys to see exactly which fields changed.
Use stable dotted keys to grep or look up a specific nested value.
To go straight from JSON to a CSV file, use JSON to CSV; to check or tidy the JSON first, use the JSON Formatter or JSON Validator.
Checked against this tool's code: it recursively walks objects and arrays, builds a dot-notation path to each leaf value and outputs pretty-printed JSON. It does not unflatten, and it uses dot indices rather than brackets.
CSV rows become a JSON array, keyed by the header row.
Flatten a JSON array of objects into CSV columns.
Turn everyday Markdown into clean HTML in your browser.
Pasted CSV becomes a real .xlsx, with leading zeros kept.
Pull the first sheet of a workbook out as plain CSV.
Each spreadsheet row becomes a JSON object keyed by the header.
The file never leaves your computer, and you don't need an account.