Is this real banking data?

No. Every amount, merchant, and account ID is randomly generated locally in your browser from sample pools. Nothing connects to a real bank and no data leaves your device.

What columns are included?

Each transaction has a transaction ID, account ID, ISO date, a signed amount, a currency code, a category, a merchant name, and a debit or credit type. Debits are negative and credits are positive.

How are amounts decided?

Amounts are random within ranges that suit each category, so groceries are small and rent or salary are large. Debits get a negative sign and credits a positive one, which keeps a running balance realistic if you sum them.

Can I get a balanced ledger?

The data is not double-entry balanced by default; it is a flat transaction list. You can sum the signed amounts to get a net cash flow, but it will not be zero unless you post offsetting entries yourself.

Is the output reproducible?

No. Each click regenerates a fresh random set. Save the output if you need a stable dataset for a demo or test fixture.

Sample Finance Dataset Generator

This tool produces realistic but completely fake financial transactions so you can build fintech dashboards, ledger viewers, and charting demos without using real banking data. Everything is generated locally in your browser.

How it works

Each row is built by picking a category, then drawing an amount from a range that fits that category:

A transaction ID like TXN-000123 and an account ID like ACC-0007.
A category (Groceries, Dining, Rent, Salary, Transfer, Utilities, …) with a realistic amount range, so a salary is large and a coffee is small.
A debit/credit flag — debits are stored as negative amounts and credits as positive, so summing the column gives net cash flow.
A merchant name appropriate to the category and an ISO date within the past year.

The result is serialized as CSV with a header row or as pretty JSON.

Tips and notes

Sum the signed amount column to demo a running-balance or cash-flow chart. Because categories carry sensible ranges, the data also works well for grouping, budgeting, and category-breakdown visualizations. Each generation is independently random, so save the file if you need a fixed fixture.