Converting JSON to CSV: How to Flatten Nested Data for Spreadsheets

JSON and CSV both represent tabular data, but they handle structure very differently. JSON can nest objects and arrays indefinitely. CSV is flat — two dimensions, rows and columns. That gap is where most conversion bugs live.

Here is a practical guide to converting JSON to CSV without losing data or your mind.

Why JSON to CSV Conversion Is Harder Than It Looks

A flat JSON array is easy. This converts in seconds:

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

id,name,role
1,Alice,admin
2,Bob,editor

The problem starts when your real-world JSON looks like this:

[
  {
    "id": 1,
    "name": "Alice",
    "address": {
      "city": "London",
      "zip": "EC1A"
    },
    "tags": ["admin", "billing"]
  }
]

Now you have choices to make. Should address.city become a column? Should tags become a comma-separated string, or multiple columns, or separate rows?

The answer depends on what you are doing with the CSV.

Approach 1: Dot-Notation Flattening

The most common approach for nested objects is to flatten them with dot notation:

id,name,address.city,address.zip,tags
1,Alice,London,EC1A,"admin,billing"

Here is a minimal JavaScript function that does this:

function flattenObject(obj, prefix = '') {
  return Object.entries(obj).reduce((acc, [key, val]) => {
    const fullKey = prefix ? `${prefix}.${key}` : key;
    if (val !== null && typeof val === 'object' && !Array.isArray(val)) {
      Object.assign(acc, flattenObject(val, fullKey));
    } else {
      acc[fullKey] = Array.isArray(val) ? val.join(';') : val;
    }
    return acc;
  }, {});
}

Then convert the array:

function jsonToCSV(data) {
  const flattened = data.map(row => flattenObject(row));
  const headers = [...new Set(flattened.flatMap(Object.keys))];

  const rows = flattened.map(row =>
    headers.map(h => {
      const val = row[h] ?? '';
      // Escape values that contain commas, quotes, or newlines
      const str = String(val);
      if (str.includes(',') || str.includes('"') || str.includes('\n')) {
        return `"${str.replace(/"/g, '""')}"`;
      }
      return str;
    }).join(',')
  );

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

The key part people forget: proper CSV escaping. A value like "Hello, world" contains a comma. Without quotes, it splits into two columns. Without doubling internal quotes, it breaks the parser.

Approach 2: Stringify Mode

Sometimes you do not need to flatten nested data. You just need it in a spreadsheet for a quick review. In that case, stringify the complex values:

function flattenShallow(obj) {
  return Object.fromEntries(
    Object.entries(obj).map(([k, v]) => [
      k,
      typeof v === 'object' && v !== null ? JSON.stringify(v) : v
    ])
  );
}

This keeps columns clean but stores {"city":"London","zip":"EC1A"} as a raw string in the cell. Readable, but you cannot sort or filter by nested properties.

Approach 3: Explode Arrays to Multiple Rows

If you have a one-to-many relationship — one user with multiple orders — you might want each array element as a separate row:

{ "userId": 1, "orders": [{"id": "A1"}, {"id": "A2"}] }

Becomes:

userId,orders.id
1,A1
1,A2

This produces more rows but is often the right shape for database imports.

Handling Edge Cases

Different objects with different keys. Real APIs return inconsistent shapes. Row 1 might have firstName, row 2 might have first_name. Your header extraction needs to union all keys, not just read the first row:

const headers = [...new Set(data.flatMap(Object.keys))];

Null and undefined values. Use ?? '' (nullish coalescing) rather than || '' — a value of 0 or false is falsy but valid and should not be replaced with an empty string.

Unicode and special characters. Emoji, accented characters, and right-to-left text all survive in CSV as long as you save the file as UTF-8. If your downstream tool opens it in Excel, Excel may interpret the encoding wrong. Adding a BOM () at the start of the file fixes this for most Excel users.

Number formats. A zip code like 00123 is a string, not the number 123. If you stringify everything, it survives. If you try to type-detect values, be careful — leading zeros mean the field is a string.

Triggering a File Download

Once you have the CSV string, downloading it from the browser is three lines:

function downloadCSV(csvString, filename = 'export.csv') {
  const blob = new Blob([csvString], { type: 'text/csv;charset=utf-8;' });
  const url = URL.createObjectURL(blob);
  const a = document.createElement('a');
  a.href = url;
  a.download = filename;
  a.click();
  URL.revokeObjectURL(url);
}

No libraries needed. Works in every modern browser.

When to Use a Library

For production use with complex data, consider Papa Parse — it handles all edge cases and is the standard CSV library for JavaScript.

For quick one-off conversions — pasting an API response and getting a CSV for a spreadsheet — you do not need a library at all.

The SnappyTools JSON to CSV Converter handles nested objects with dot-notation flattening, stringify mode for complex values, live CSV preview, and file download. It runs entirely in your browser with nothing uploaded.

Summary

• Flat JSON arrays: trivial to convert
• Nested objects: flatten with dot notation or stringify
• Arrays of values: join with separator or explode to rows
• Always escape commas, quotes, and newlines in cell values
• Union all keys across all rows, not just the first row
• For browser downloads, use Blob + URL.createObjectURL — no dependencies needed