Account

What new tool do you want?

Generators

Fake Test Data Generator (JSON / CSV)

Generate realistic fake people — names, emails, phones, addresses, companies — as JSON or CSV for testing and demos.

How to use

  1. 1Choose how many records and which fields to include.
  2. 2Pick JSON or CSV, then copy or download the data.

Frequently asked questions

Is this data real?

No. Names are combined at random, emails use the reserved example.com/.org/.net domains and phones use the fictional 555-01xx range, so nothing reaches a real person.

Can I use it in automated tests?

Yes. Download the JSON as a fixture file. For repeatable tests, generate once and commit the file rather than regenerating.

Are the addresses valid?

City, state and ZIP combinations are real, but street numbers and names are random — don't use them for address validation tests.

Why nothing here can reach a real person

Emails are built as `firstname.lastname + random 1–99 digits` at one of three domains reserved specifically for documentation and testing by IANA: example.com, example.org and example.net — mail sent there bounces or goes nowhere, by design, not by luck. Phone numbers use the North American `555-01xx` range, which the US numbering plan permanently reserves for fictional use in film, TV and software (only 555-0100 through 555-0199 are reserved this way; the rest of the 555 range is in normal service, which is why the generator sticks specifically to 01xx). Addresses pair a real city/state/ZIP combination with a randomly generated street number and name, so the geography is plausible but the exact address isn't a real mailing location.

What's realistic vs. what's just random

City, state and ZIP come from a curated list of actual US locations, so those three fields are internally consistent and geographically real. Everything layered on top of that — the house number, street name, birth date, username — is independently randomized and has no relationship to the city chosen, which is why the generator's own FAQ is upfront that these records aren't suitable for address-validation testing; they're meant to exercise your code's handling of realistic-shaped data, not to represent real deliverable addresses.

JSON vs CSV, and why flattening happens

JSON output preserves the natural nested shape — `address` is an object with its own `street`, `city`, `state`, `zip` and `country` keys. CSV can't represent nesting, so the export step flattens any object field into separate columns with an underscore-joined prefix (`address_street`, `address_city`, and so on) before writing the header row and one row per record. If you toggle off a field entirely (like Company), it simply won't appear as a column — the CSV header always matches exactly whichever fields you left enabled.

Worked example: Fake Test Data Generator

Requesting 3 records with Name, Email and Phone enabled (and everything else off) as CSV produces a header of `firstName,lastName,email,phone` followed by three rows such as `Maya,Chen,[email protected],+1-312-555-0187` — a realistic-looking fixture you can commit directly into a test suite, safe in the knowledge that the email domain and phone range both resolve to nobody.

Switching every field off still returns a name

Records start at 10 and are kept between 1 and 1,000. If every field toggle is off, a name is still included, so you never get an empty row. Phone numbers use a reserved 555 exchange and end in 01 followed by two digits, and email addresses use only example.com, example.org, or example.net. JSON keeps an address as a nested group, indented with two spaces. CSV cannot nest, so those pieces become separate columns such as address_street. Username starts switched off; the other groups start on. The header row contains only the groups you left on, so turning off id, address, company, birth date, and phone leaves firstName, lastName, and email, and three records then mean four lines once the header is counted.

Related tools

More: Generators