Test Data Generator — JSON, CSV, SQL
This free test data generator produces realistic fake records — names, email addresses, words, numbers, and UUIDs — in JSON, CSV, or SQL INSERT format. Define a simple schema, choose how many rows you need (up to 1,000), and get structured sample data for seeding databases, populating demos, writing tests, and prototyping APIs.
The data is powered by the Faker library and generated entirely in your browser. Because every value is synthetic, you can build convincing demos without touching real customer data — an easy win for privacy and GDPR compliance.
How to use Test Data Generator
- Define the schema as comma-separated name:type pairs, e.g. name:name, email:email, age:number.
- Available field types: string (random word), number, email, name (full name), and uuid.
- Choose the output format — JSON, CSV, or SQL — and the number of records.
- Press Generate Data, then copy the result into your fixture file, spreadsheet, or database client.
Output formats
JSON emits an array of objects for API mocks and fixtures; CSV emits a header row plus one line per record for spreadsheets and imports; SQL emits a single INSERT INTO my_table statement with quoted values ready to run against a test database.
Schema: name:name, email:email, age:number
JSON → [ { "name": "Jane Cooper", "email": "jane@example.net", "age": 42 } ]
CSV → name,email,age\nJane Cooper,jane@example.net,42
SQL → INSERT INTO my_table (name, email, age) VALUES ('Jane Cooper', 'jane@example.net', '42');Related terminology
- Fixture
- A fixed set of sample data loaded before tests run so results are predictable and repeatable.
- Seeding
- Populating a fresh database with initial or sample rows — exactly what the SQL output is for.
- Faker
- The widely used open-source library that synthesizes realistic names, emails, and other fake values.
- Synthetic data
- Artificially generated records that mirror real data's shape without exposing any real person's information.
Frequently asked questions
- How do I define the schema?
- Write comma-separated field definitions in the form fieldName:type. For example id:uuid, name:name, email:email, score:number produces four columns with matching fake values.
- Is the generated data based on real people?
- No. Names, emails, and other values are synthesized by the Faker library; any resemblance to real people is coincidental. That is what makes it safe for demos and screenshots.
- Can I use the SQL output directly?
- Yes — it is a standard INSERT statement (table name my_table; rename as needed). String values are quoted and escaped. Note that numeric values are also quoted, which the major databases accept and convert.
- How many records can I generate?
- Up to 1,000 per run, which is plenty for fixtures and demos. For larger volumes, run several batches or use Faker directly in a script.