Developer & Data100% Free for AllClient-Side PrivateAuto-saved (3 days)

Sample CSV & JSON Generator

Create test data files at any size — CSV with headers and rows, JSON arrays of records, or plain text. Useful for testing importers, parsers, and upload limits.

File type
Target size
KB

Testing an importer means having data to import, and hand-writing a thousand rows of CSV is nobody's idea of a good afternoon. This generates plausible records — names, emails, cities, amounts, dates — as CSV, JSON, or plain text, at exactly the file size you need.

How to generate sample data files

  1. Choose a format: CSV for tabular imports, JSON for API fixtures, plain text for anything else.
  2. Set a target size. Larger files simply contain more records.
  3. Generate four samples and inspect the filenames and sizes.
  4. Download the one you need.

The Sample CSV & JSON Generator runs entirely in your browser — nothing you enter is uploaded, stored, or logged.

When to use this tool

Testing a CSV importer

Importers behave differently on ten rows and ten thousand. Generating both sizes shows whether batching, progress reporting, and timeouts hold up.

Seeding fixtures for development

A JSON array of a few hundred records is enough to make a list view, pagination, and search feel realistic before a real API exists.

Checking upload and parse limits

A 10MB CSV is a very different proposition from a 10KB one for a browser-side parser. Having both makes the limits measurable rather than guessed.

Things worth knowing

  • CSV output includes a header row, so it imports cleanly into spreadsheets and most database tools.
  • JSON is pretty-printed, which makes it larger than minified output but far easier to read while debugging.
  • Records use plausible values — real-looking names, emails, cities, and amounts — rather than repeated filler.
  • Sizes are exact, because text formats can be trimmed to the byte.

Frequently Asked Questions

Is the data realistic?

It uses plausible names, emails, cities, and amounts — enough to exercise a parser or importer.

Is the data realistic enough to test with?

For structure, yes — records have consistent fields, valid-looking emails, real city names, and varied numeric and date values, which is enough to exercise a parser, importer, or table view. It is not statistically realistic, so it is not a substitute for real data when testing analytics.

How many rows will I get?

As many as fit the size you asked for. A CSV row here is roughly 70 bytes, so 1MB is around 15,000 rows. The file is filled with records and then trimmed to the exact byte count.

Can I choose the columns or fields?

Not currently — the schema is fixed at id, name, email, city, amount, and created_at, which covers the common shapes of test data. For a custom schema, generate a file here and edit the header and a row, or script it.

Does the data contain anything real?

No. Every value is generated, the email addresses use the reserved example.com domain, and nothing corresponds to a real person. It is safe to commit to a repository or share.