Create fake test data in CSV format for your projects.
About Generate Test Data
Developers, testers and trainers constantly need realistic but fake records: a list of users to seed a database, a CSV to test an import, or a dataset for a demo where real customer data cannot be shown. This generator builds rows from field types such as full name, email, UUID, random number, city, country and date, and exports them as CSV or JSON.
Define the fields, choose how many rows you need, click Generate and export. Everything is produced by JavaScript in your browser, so it works offline and no data is stored.
How to use Generate Test Data step by step
Edit the list of fields: each has a name (used as column header) and a type. Add fields with "Add field" or remove them with the trash icon.
Set the number of rows to generate (10 by default).
Click "Generate data". A preview table shows the result.
Click "Export CSV" to download datos_de_prueba.csv or "Export JSON" for datos_de_prueba.json.
Input and output example
Input
Fields: id (UUID), nombre_completo (Full name), email (Email)
Rows: 2
100% free and online: no payments, no software or extensions to install.
Privacy-first: processing happens in your browser, your data never leaves your device.
No sign-up: paste your data and get the result in one click.
Fast and reliable: handle small or large files in seconds, ready for Excel and other tools.
Common use cases
Seed a development or staging database with a few hundred users before a demo.
Test the CSV import of a CRM, e-commerce or accounting application without real customers.
Create sample data for a spreadsheet tutorial, a training workshop or a screenshot.
Produce fixtures for automated tests in Cypress, Playwright or pytest.
Limits and tips
Names, cities and countries come from small Spanish-language lists (Juan, María, Madrid, España...), so the data looks Spanish. Emails are built from the names with the domain ejemplo.com.
Random numbers are integers between 0 and 9999; dates are ISO strings (YYYY-MM-DD) between 2000-01-01 and today; UUIDs are version 4.
Names and emails are generated independently, so the email in a row does not match the name in the same row.
In the CSV export every value is quoted; the JSON export is an array of objects with two-space indentation.
Frequently asked questions
What field types are available?
Full name, first name, last name, email, random number, UUID, city, country and random date. Each row picks values at random for every field, and the same value can appear in several rows.
Is the generated data realistic?
It is plausible but deliberately simple: names are drawn from a short list of common Spanish names and surnames, cities are large Spanish cities and countries are Spanish-speaking. It is enough to test layouts and imports, but not to simulate a real population.
What is the difference between test data and anonymised data?
Test data is invented from scratch and has no link to real people, so it can be shared freely. Anonymised data starts from real records and removes or masks identifiers, which is harder to do safely. For demos and automated tests, generated data avoids privacy risks entirely.
Can I generate thousands of rows?
Yes. Generation is fast, but the preview renders every row, so very large sets make the page slow; export right after generating. For millions of rows a script with a library such as Faker is more practical.
Is anything sent to a server?
No. The rows are generated and exported entirely in your browser.