Load Test Data Generator

Runs in browser Performance

Produce CSV/JSON feeders with thousands of unique users, credentials, ids and payload variants, formatted for popular load-testing tools. Synthetic test data only.

Privacy: This tool runs entirely in your browser. Your input never leaves your device.

How to use Load Test Data Generator

  1. Describe one column per line, e.g. "username:username", "amount:amount:5-500", "plan:pick:free|pro".
  2. Choose the number of rows, the target tool and a file name; add a seed for reproducible data.
  3. Click "Generate" and review the preview table.
  4. Copy the loader snippet into your script and download the feeder file next to it.

Load Test Data Generator features

  • Column DSL: name:type[:options] with 20 generator types (sequence, uuid, username, token, amount ranges, pick lists…)
  • Up to 20,000 unique rows, reproducible with a seed
  • Formats for k6 (JSON + SharedArray), JMeter (CSV Data Set Config), Gatling and Locust feeders, plus plain CSV/JSON
  • Loader snippet generated for the selected tool
  • Preview table and dataset statistics before download
  • Synthetic values only — example.com emails, 555 phone numbers, fictional names

Load Test Data Generator example

k6 feeder with credentials and plans

Input:

username:username
password:password:12
plan:pick:free|pro|enterprise
Rows: 3 · Format: k6 · Seed: demo

Output:

[
  { "username": "layla0001", "password": "Pw!k3Qz9…", "plan": "pro" },
  { "username": "omar0002", "password": "Pw!Ab7Xc…", "plan": "free" },
  { "username": "noor0003", "password": "Pw!Zt2Lm…", "plan": "enterprise" }
]

const users = new SharedArray('users', () => JSON.parse(open('./users.json')));

Frequently asked questions about Load Test Data Generator

Which column types are available?

sequence, uuid, username, password, email, first/last/full name, token, int and float ranges, amount, boolean, date, datetime, phone, pick (a|b|c), string, json and constant.

How do I get the same data twice?

Set a seed. The generator uses a deterministic PRNG, so the same columns, row count and seed always produce the same rows — handy for reproducible test runs.

Which format should I pick for my tool?

k6 gets JSON plus a SharedArray snippet; JMeter, Gatling and Locust get CSV with the matching loader snippet; CSV/JSON are generic.

Is the data safe to use?

Everything is synthetic: names come from a fixed list with obviously fake surnames, emails use example.com and phone numbers use the reserved 555 range.

How many rows can I generate?

Up to 20,000 rows in the browser. For millions of rows, generate a seed-based dataset on the load generator itself.

Technical notes

Rows are produced by a seeded mulberry32 PRNG shared with the other test-data tools, so a seed yields identical output across sessions and machines. Without a seed the tool uses fresh randomness on every run and "Regenerate".

Usernames combine a fictional first name with a zero-padded index and are de-duplicated; sequence columns count up from the start value so every row has a unique identifier for correlation in results.