What are Velocity Checks?
A velocity check is a monitoring control that flags when the number or value of transactions within a set period exceeds an expected limit.
Velocity checks measure how fast money moves through an account. A sudden burst of activity, such as dozens of transfers in an hour or a large inflow that leaves again within minutes, often signals money laundering, money mule activity, or fraud.
Velocity checks are one of the core rule types in transaction monitoring, alongside threshold and pattern rules. They are simple to set up and easy to explain, which makes them a foundation of most AML and fraud programs.
What are examples of velocity checks?
Velocity checks can track counts, amounts, or both over a defined window. Common examples include the following.
- More than five wire transfers in 24 hours
- Total cash deposits above a set amount within one day
- Several payments to new beneficiaries within a short period
- Funds received and sent out again within minutes or hours
- Multiple accounts opened or logins attempted from the same device in a short time
How do velocity checks work?
Velocity checks add up activity across a rolling time window and compare the total against a limit. AML teams often use windows of 1, 7, 30, or 90 days to spot patterns that build over time, such as structuring. Real-time fraud and payment controls use much shorter windows, such as the last 5, 30, or 300 minutes, so they can act before a payment settles.
Velocity checks work best when they compare activity to the customer's own baseline. A spike measured against how a customer usually behaves catches more genuine risk, and creates fewer false positives, than a single fixed limit applied to everyone.
How do firms tune velocity checks?
Firms tune velocity checks by testing them against historical data and adjusting the limits and time windows. Window length matters because legitimate patterns like payroll cycles can trigger alerts if the window is set poorly. Many firms also apply tighter velocity limits to high-risk customers and looser limits to low-risk segments, then review results in regular calibration meetings.
