JSON to CSV Converter

What is CSV and Why Convert from JSON?

// Input JSON
[
  { "id": 1, "name": "Alice", "role": "admin" },
  { "id": 2, "name": "Bob", "role": "editor" }
]

// Output CSV (RFC 4180 preset: comma + CRLF + no BOM)
id,name,role
1,Alice,admin
2,Bob,editor

// Same input with Stringify mode + nested data
[
  { "id": 1, "tags": ["a", "b"] }
]

// Becomes
id,tags
1,"[""a"",""b""]"

Key Features

Examples

[{"id":1,"name":"Alice Chen","email":"alice@example.com","role":"admin"},{"id":2,"name":"Bob Garcia","email":"bob@example.com","role":"editor"},{"id":3,"name":"Carol Singh","email":"carol@example.com","role":"viewer"},{"id":4,"name":"Dan Müller","email":"dan@example.com","role":"viewer"}]

[{"id":1001,"title":"Bug: login redirects to 404","state":"open","labels":["bug","priority:high"],"user":{"login":"alice"}},{"id":1002,"title":"Feature: dark mode toggle","state":"open","labels":["enhancement"],"user":{"login":"bob"}},{"id":1003,"title":"Docs: update README","state":"closed","labels":["docs","good-first-issue"],"user":{"login":"carol"}}]

[{"_id":{"$oid":"6634a1b2c3d4e5f600000001"},"email":"alice@example.com","metadata":{"signupDate":"2026-01-15T10:30:00Z","preferences":{"newsletter":true,"theme":"dark"}}},{"_id":{"$oid":"6634a1b2c3d4e5f600000002"},"email":"bob@example.com","metadata":{"signupDate":"2026-02-20T14:15:00Z","preferences":{"newsletter":false,"theme":"light"}}}]

{"event":"signup","user":"alice","ts":1715000000}
{"event":"login","user":"alice","ts":1715000060}
{"event":"checkout","user":"alice","ts":1715000300}

[{"id":"ord-001","customer":{"name":"Alice Chen","email":"alice@example.com","address":{"city":"Seattle","country":"US"}},"items":[{"sku":"SKU-100","qty":2},{"sku":"SKU-205","qty":1}]},{"id":"ord-002","customer":{"name":"Bob Garcia","email":"bob@example.com","address":{"city":"Madrid","country":"ES"}},"items":[{"sku":"SKU-100","qty":1}]}]

[{"name":"GET /users returns 200","status":"pass","duration":142},{"name":"POST /users creates record","status":"pass","duration":287},{"name":"GET /users/999 returns 404","status":"fail","duration":98,"error":"Expected 404, got 500"},{"name":"DELETE /users/1 returns 204","status":"pass","duration":156}]

How to Use

1. 1

   Paste your JSON

   Enter or paste your JSON into the input field above. The tool accepts arrays of objects, single objects, and NDJSON (line-delimited JSON) — auto-detected. You can also click 'Load example' to try a sample like a REST API response, MongoDB export, or NDJSON event log.

2. 2

   Pick a preset (or tweak Options)

   Click RFC 4180 (default), Excel (EU semicolon + BOM), TSV (Tab), or Pipe to dial in a target format in one click. Open the Options panel for fine control: Delimiter, Header, Quote (Auto/Always), Line ending (LF/CRLF), BOM, and Nested (Flatten/Stringify).

3. 3

   Copy or Download the CSV

   Click Copy to grab the CSV to your clipboard, or Download to save it as a .csv (.tsv with TSV preset) file ready for Excel, Google Sheets, PostgreSQL COPY, or any data pipeline. For round-trips, click Swap direction to convert CSV back to JSON in place.

Common Conversion Pitfalls

42

[{"value": 42}]

[
  {"name": "GET /users", "status": "pass"},
  {"name": "GET /users/999", "status": "fail", "error": "500"}
]

// Output CSV (note empty cell in row 1)
name,status,error
GET /users,pass,
GET /users/999,fail,500

{"id": 9007199254740993}
// JavaScript reads as 9007199254740992 (precision lost)

{"id": "9007199254740993"}
// CSV preserves the string; Excel displays exactly

name,role
Smith, Jr.,admin
// Parser reads 3 columns: "Smith", " Jr.", "admin"

name,role
"Smith, Jr.",admin
// Parser reads 2 columns: "Smith, Jr.", "admin"

id,name,price
1,Alice,1,234.56
// Excel-EU mis-parses 1,234.56 as two columns

// Excel preset output: ; + CRLF + BOM
id;name;price
1;Alice;1234,56
// Excel-EU reads cleanly with comma decimal

[
  {"event": "signup"},
  {"event": "login"}
]
// Detected as a regular JSON array (works, but not NDJSON path)

{"event": "signup"}
{"event": "login"}
// Each line is one JSON value — auto-detected as NDJSON

Common Use Cases

API Response to Spreadsheet

Paste a REST API response (array of user/order/event objects) and get a clean tabular CSV ready for Excel, Google Sheets, or Numbers. The most common use case — analysts and PMs need spreadsheet-shaped data, engineers ship JSON from the backend.

MongoDB Export to Data Warehouse

Convert mongoexport JSON output (with $oid wrappers and nested metadata documents) into CSV for loading into BigQuery, Snowflake, or Redshift. Stringify mode preserves the nested shape losslessly when downstream tools support JSON-in-cell.

GitHub Issues Triage

Pull issues from /repos/{owner}/{repo}/issues, paste the JSON, and get a flat CSV with id, title, state, labels, and user.login as columns. Drop into Sheets to filter, sort, and assign during sprint planning.

NDJSON Event Log Review

Streaming logs from Cloud Logging, Datadog, Vector, or your own pipeline often arrive as NDJSON (.ndjson, .jsonl). Paste the file contents directly — auto-detected — and get a CSV for ad-hoc analysis without spinning up a real ETL pipeline.

E-commerce Order Extraction

Convert nested order JSON (customer.address.city, items array) to a flat CSV for finance, fulfillment, or fraud review. Flatten mode produces one column per leaf field, ready to load into Excel for ad-hoc reporting.

Postman/Newman Test Report

Postman test exports include mixed-shape rows (some with optional error fields). Paste the JSON, get a CSV with all keys unioned and missing cells filled — perfect for sharing failed-test reports with non-engineers in Sheets.

Technical Details

RFC 4180 State-Machine Parser

Both directions use a proper finite-state-machine parser following RFC 4180. States include UnquotedField, QuotedField, AfterQuote, RowEnd, and EndOfInput. The parser correctly handles quoted fields containing the delimiter, embedded CR/LF inside quoted fields, escaped double quotes (doubled, like ""), and trailing newlines. This produces output that round-trips losslessly through Python's csv module, PostgreSQL COPY, AWS S3 SELECT, and any compliant parser.

Big-Integer Detection Algorithm

Before passing JSON through JSON.parse() (which would silently round large integers via IEEE 754), the tool runs a regex scan over the raw JSON text looking for integer literals outside the safe range (-2^53+1 to 2^53-1). When detected, the affected field paths are recorded and a Big-integer warning banner appears below the output. The CSV writer then preserves these values as strings, ensuring exact precision through Excel, Google Sheets, and any text-aware downstream consumer.

BOM and UTF-8 Encoding for Excel

All input and output is UTF-8. The optional UTF-8 BOM (0xEF 0xBB 0xBF) is prepended to the output when the BOM toggle is on or when the Excel preset is selected. Excel on Windows uses the BOM to detect UTF-8 encoding — without it, Excel falls back to the system locale (Windows-1252 in the US, Windows-1251 in Russia, etc.) and mangles non-ASCII characters. Modern parsers (Python csv, Pandas, jq, PostgreSQL) generally do not need the BOM and may include it as a stray character at the start of the first cell, so leave BOM off for non-Excel pipelines.

Best Practices

Choose a Preset Before Tweaking Options

RFC 4180, Excel, TSV, and Pipe presets dial in five options at once (delimiter, line ending, quote mode, BOM, header). Pick the closest preset first, then tweak individual options only if needed — this avoids the common mistake of flipping one option and forgetting another (e.g., switching to semicolon but leaving LF, which Excel-EU on Windows still mis-parses).

Flatten for Analysis, Stringify for Round-Trips

Use Flatten when the destination is Excel, Sheets, or a one-time analysis — dotted keys produce the cleanest spreadsheet layout. Use Stringify when you need a CSV → JSON → CSV round-trip without data loss — arrays and objects survive as JSON inside a single cell. Switching mid-export and re-running is cheap; pick based on the consumer.

Use BOM Only for Excel

The UTF-8 BOM is required for Excel on Windows to detect encoding correctly. Every other parser (Python csv, Pandas, jq, PostgreSQL COPY, BigQuery) either ignores the BOM or includes it as a stray character at the start of the first cell, breaking column names. Leave BOM off for pipelines and toggle on (or use the Excel preset) only when the destination is Excel.

Keep Large IDs as Strings in JSON

Twitter snowflake IDs, Discord IDs, MongoDB Long fields, and K8s resourceVersion are 64-bit integers that exceed JavaScript's safe range (2^53 - 1). Store them as JSON strings ("id": "9007199254740993") before converting — the CSV will preserve the digits exactly, while a numeric literal would be silently rounded by JSON.parse().

Validate Row Shape Before Loading

Mixed-shape rows (some objects with keys others don't) are merged with empty cells in the output, and the tool shows a Schema notes warning. For strict-schema consumers (BigQuery, Redshift COPY), validate that all rows share the same keys before exporting — or handle the missing values explicitly in your pipeline. Use our JSON Formatter to inspect the input shape first.

Frequently Asked Questions