Simulated data imitates how real records behave, closely enough to exercise your application: names match emails, ages match birthdates, orders have plausible prices and statuses, and customers have different numbers of orders. RNDGen'sdata generator does this out of the box.
Fake, simulated or synthetic?
| Fake data | Simulated data | |
|---|---|---|
| Each value | Random of the right type | Realistic and within the ranges you set |
| Fields in a row | Independent | Agree: email from name, age from birthdate, ZIP from state |
| Rows across tables | No relation | Children point to real parents, a varying number per parent |
| Good for | Filling a form or a table quickly | Testing logic, reports, imports and performance |
Both are synthetic data: generated, not collected, and free of real personal information.
How RNDGen simulates real data
- Consistent rows: person, address and company fields are generated together, so they match.
- Related tables: users, their orders and order lines, with a minimum and maximum of children per parent.
- Your own value lists: statuses, plans and categories with the Custom list field; repeat a value to make it more frequent.
- Ranges: numbers, prices, dates and ages within the limits you set.
- Blanks on purpose: a share of empty values per field, to test how your code handles missing data.
- 41 languages and regions with a free account and 54 on Pro and Team, so a German customer gets a German address and phone number.
- Reproducible: a seed and a reference date give the same dataset every time.
When to use it
Load and performance tests that need millions of believable rows, reports and dashboards that need realistic distributions, demo environments that should look lived-in, and import or migration tests. Start from a template or letAI design the schema from a description.
FAQ
What is the difference between fake, simulated and synthetic data?
Fake data fills each field on its own. Simulated data also follows the rules of the real world: fields agree, rows relate to other rows, values fall in realistic ranges. Synthetic data is the umbrella term for any data generated instead of collected; sometimes it means data that reproduces the statistics of a real dataset.
Can I control the distribution of values?
Yes, within each field: number and date ranges, the share of true values, the share of blanks, and weighted value lists (repeat a value in a custom list to make it more common).
Can child rows vary per parent?
Yes. For a related table you set the minimum and maximum number of child rows per parent, so some customers have no orders and some have many.
Will I get the same simulation again?
Yes. The seed and the reference date fix every value, so the same schema produces the same dataset on any day.