Velocity of change counts how many data elements differ between a new transaction and previous ones from the same identity — a different signal from velocity of use, which counts repetition. This one is built to catch data that's shifting, not staying the same.
What it catches
Identity morphing attempts, stolen card numbers in active use, and multiple attacks from the same perpetrator — tracked across card numbers, billing and shipping addresses, phone numbers, email addresses, and account numbers.
Where this typically lives
Most major fraud-screening platforms already include this. When evaluating one, confirm whether custom fields can be added and whether it tracks changes at the account level, not just the transaction level.
Building it yourself
It needs database support and two calls — one to increment, one to retrieve — the same pattern as velocity of use. Cost stays low if you already have the database infrastructure and integrated order processing in place; many ERP systems and decision servers already include this capability without extra licensing.
Applying velocity of change to new account openings specifically — not just orders — helps catch a fraudster trying to spin up multiple accounts with slightly varied data before ever placing an order.