Is any of this data real?

No. Every value is randomly assembled from generic name and street word-lists and synthetic number formats. The data is structurally valid (correct formats and check digits) but does not correspond to any real person, account, or address.

Are the IBANs and NI numbers valid?

They follow the correct format and pass the relevant check — the IBAN passes the ISO 13616 MOD-97 checksum, and the National Insurance number avoids the disallowed prefixes and suffixes. They are format-valid placeholders, not issued to anyone.

What is the seed for?

Seeding makes generation deterministic: the same seed and count always produce the same records. This lets test suites assert against stable fixtures and lets teammates reproduce the exact dataset you used.

Can I use this for production data?

No. It is intended for testing, QA, demos, and documentation. Because the data is fictional it is safe to commit to repositories and share in screenshots without leaking real PII.

Does anything leave my browser?

No. Generation runs entirely client-side with a small pseudo-random generator. There are no network requests, so the fake records are never transmitted or logged anywhere.

Fake Data & Test PII Generator

Fictional identities for safe testing

When you need to populate a test database, take a clean demo screenshot, or write a fixture, using real personal data is a privacy and compliance risk. This generator builds entirely fictional but well-formed identities — names, dates of birth, addresses, National Insurance numbers, IBANs, phone numbers, and emails — that look real enough for realistic testing while corresponding to nobody.

How it works

A small seedable pseudo-random number generator (a mulberry32-style hash) drives all selection. Given a seed, the stream of random values is deterministic, so the same seed and count regenerate identical records — perfect for reproducible fixtures. With an empty seed a time-based seed is used for fresh data each run.

Each field is constructed to be format-correct:

National Insurance number follows the AA 00 00 00 A pattern, skips the prefixes HMRC never issues (D, F, I, Q, U, V as first letter; O as second) and uses a valid A–D suffix.
IBAN is generated for a sample country, the check digits are computed with the real ISO 13616 MOD-97 algorithm so the number passes IBAN validators.
Email is derived from the generated name plus a sample domain; phone uses a country dialling format; address is assembled from street/town word-lists.

Tips and notes

Use a fixed seed in automated tests so assertions stay stable across runs, and an empty seed when you just want a fresh batch for a demo. Because every value is synthetic, the output is safe to commit to source control and share publicly. Do not rely on the generated NI numbers or IBANs as if they belong to anyone — they are deliberately fictional placeholders that merely satisfy format checks.