JSON to CSV Conversion Guide — Manual Code and Libraries

An API returns JSON, but the person who needs the data wants an Excel file. Converting a JSON array of objects into CSV is a common task for exports, reports, and data handoffs. This guide covers a manual JavaScript implementation you fully control, plus the libraries most teams reach for in production.

The Basic Shape of the Problem

// JSON input
[
  { "id": 1, "name": "Priya Sharma", "city": "Bengaluru", "total": 1499 },
  { "id": 2, "name": "Rahul Verma", "city": "Pune", "total": 899 }
]

// CSV output
id,name,city,total
1,Priya Sharma,Bengaluru,1499
2,Rahul Verma,Pune,899

A Manual JSON-to-CSV Function

This handles the tricky part correctly: escaping values that contain commas, quotes, or newlines, per the CSV spec (RFC 4180):

function jsonToCsv(records) {
  if (records.length === 0) return '';

  // Collect the union of all keys across every record
  const columns = [...new Set(records.flatMap(r => Object.keys(r)))];

  const escapeCell = (value) => {
    if (value === null || value === undefined) return '';
    const str = String(value);
    // Quote the cell if it contains a comma, quote, or newline
    if (/[",\n]/.test(str)) {
      return '"' + str.replace(/"/g, '""') + '"';
    }
    return str;
  };

  const header = columns.map(escapeCell).join(',');
  const rows = records.map(record =>
    columns.map(col => escapeCell(record[col])).join(',')
  );

  return [header, ...rows].join('\n');
}

const orders = [
  { id: 1, name: "Priya Sharma", note: "Deliver, please call first" },
  { id: 2, name: "Rahul Verma", note: null }
];

console.log(jsonToCsv(orders));
// id,name,note
// 1,Priya Sharma,"Deliver, please call first"
// 2,Rahul Verma,

Handling Nested Objects

CSV has no concept of nested structures, so nested objects must be flattened first — see our flattening guide for the recursive function. A quick example:

const nested = [
  { id: 1, customer: { name: "Priya", city: "Bengaluru" } }
];

// Flatten first: { id: 1, "customer.name": "Priya", "customer.city": "Bengaluru" }
// Then run jsonToCsv() on the flattened records.

Using a Library Instead

• json2csv (Node.js) — handles nested fields, custom delimiters, and streaming for large datasets.
• papaparse (browser + Node.js) — widely used for both parsing and generating CSV, works well client-side too.
• pandas (Python) — pd.DataFrame(data).to_csv('output.csv', index=False) converts JSON-derived data to CSV in one line.
• csv-writer (Node.js) — simple, promise-based CSV writing for Node.js scripts.

# Python example with pandas
import pandas as pd
import json

with open('orders.json') as f:
    data = json.load(f)

df = pd.DataFrame(data)
df.to_csv('orders.csv', index=False)

Common Pitfalls When Converting JSON to CSV

• Inconsistent fields across records — always compute the full column set first, not just the first record's keys.
• Commas and quotes in values — unescaped values silently corrupt the CSV structure; always quote and escape properly.
• Numbers with leading zeros — Excel may strip leading zeros from values like PIN codes or account numbers unless the cell is explicitly quoted as text.
• UTF-8 encoding — save the output file with a UTF-8 BOM if it needs to open correctly in Excel with Indian language characters.
• Arrays inside a field — decide upfront whether to join them with a delimiter (e.g. semicolon) or expand them into separate rows.

Frequently Asked Questions