Skip to content
DuckConvert Mascot
DuckConvert
Blog

Mock Data Generator

Define a schema, pick a type per column, and get up to 1,000 rows of placeholder JSON or CSV. Nothing it produces refers to a real person. Runs in your browser.

Schema Definition

Define the fields for your mock data.

Generation Settings

Output

Your generated mock data.

Click Generate to create data...

Streamlined Database Testing

Modern databases and APIs require robust and diverse dataset inputs to simulate edge cases, benchmark query times, and validate input types. Our schema builder supports core data types including UUIDs (RFC-4122 compliant), emails, random integers, booleans, formatted ISO dates, and alphanumeric strings. Designing local mock environments has never been easier or faster.

Enterprise Security & Data Privacy

  • Zero Server Logging: No network packets containing your test structures ever traverse the internet.
  • Custom Schema Setup: Add, edit, or remove fields dynamically. Set custom keys to match your database field naming conventions.
  • Instant Clipboard & File Export: Copy results directly with one click, or trigger a clean, locally compiled file download.

What each field type produces

There are six types, and it is worth knowing exactly what comes out of each before you build a test around it.

  • String is a numbered placeholder — Value 1, Value 2, and so on. It is not a fake-name generator, and deliberately so: plausible invented names are the ones that get pasted into a staging system and mistaken for real records.
  • Email follows the same pattern at [email protected]. example.com is reserved by RFC 2606 precisely so that it can never belong to anyone, so a test harness that accidentally sends mail to this data cannot reach a real inbox.
  • Number is a random integer from 0 to 999.
  • Boolean is an even coin flip.
  • UUID comes from the browser’s own crypto.randomUUID(), so these are real version-4 UUIDs from a cryptographic random source and will not collide in practice.
  • Date is a calendar date in YYYY-MM-DD form, drawn uniformly from roughly the last four months. Always in the past, never in the future — if you are testing expiry logic or date sorting across a wider span, this will not give you the range you need.

The output is not reproducible

There is no seed field, and that is a real limitation rather than a missing label. Press Generate twice with the same schema and the numbers, booleans, dates and UUIDs all change; only the string and email columns, which are derived from the row index, come back the same. If your test needs to assert against a fixed fixture, generate once and commit the file rather than regenerating it in the test.

What it is not

This is placeholder data for filling a table, checking a CSV importer, or seeing how a layout behaves with a thousand rows in it. It is not a statistical simulator: the columns are independent of one another, so nothing correlates the way real records do, and distributions are uniform rather than shaped like anything you would measure in the world. Load-testing a query planner on this data will tell you very little about production.

It is also not an anonymiser. It invents rows from nothing; it does not take a real dataset and mask it. If you need to preserve the shape of production data while removing the people from it, that is a different problem with different tools.

Practical limits

  • 1,000 rows per run. Everything is built in memory and rendered into a preview pane, so the ceiling keeps the page responsive. Generate repeatedly and concatenate if you need more.
  • The CSV is quoted to RFC 4180. Name a column city, country and it comes back as one column rather than two, because a field containing a comma, a quote or a line break is wrapped and any embedded quote doubled. Rows end with CRLF, which is what the specification asks for and what spreadsheets expect.
  • JSON output is a flat array of objects. There is no nesting, no arrays inside rows, and no relationships between generated tables.

Mock Data Generation Frequently Asked Questions

What is a Mock Data Generator?

A Mock Data Generator is a utility used by software developers, database administrators, and QA engineers to create realistic placeholder datasets. It simulates structured production databases, allowing teams to test application logic, API performance, database schemas, and data pipelines without exposing actual user information.

Where is the generated mock data processed?

DuckConvert's Mock Data Generator operates 100% locally on your computer. All fields, record counts, and export streams are compiled directly in your web browser session using standard client-side JavaScript. Absolutely zero data is sent to external servers, making it 100% compliant with enterprise security and NDA policies.

What formats can I export my simulated datasets in?

JSON or CSV. JSON suits fixtures for services, document stores and REST endpoints, since it carries nested structure and types. CSV is flat but reads straight into Excel, Google Sheets and a SQL bulk import (PostgreSQL, MySQL, SQLite).

Is there a row-generation limit?

Because all generation logic runs on your local processor, the performance is limited only by your browser's allocated memory (RAM). While you can generate thousands of records, we recommend keeping batch runs up to 10,000 rows at a time to ensure instant, fluid browser performance without script lag.

Other data & developer utilities tools

All Data & Developer Utilities