A velocity check is used to try to find fraud signals when there are a lot of payments being run through an account. There are patterns that can be picked up on automatically that are common signs that there may be fraudulent activity happening.
Velocity Checks and Fraud Prevention
This is the system that picks up on any unusual transactions happening on your merchant account and will cause it to freeze until you can confirm that the transactions were in fact you and not an attempt at fraud. To find these patterns a system checks for patterns over a short period of time to look for unusual activity.
There are a few ways to check for this unusual activity, using IP addresses, a device identifier, a number on a card or the payment method used.
The system will also check for things like how much has been spent in the last day compared to the rest of the year, and how many times a card has been used in a short period to identify if the card is being abused by someone that isn’t the owner.
These are called velocity rules and are used by financial systems to check for spending activity that is outside of the norm.
An example of when this may be triggered is when several large payments come out of an account all in one day. This is a clear signal that something abnormal is happening on the account that should be flagged at potentially abnormal, and not the usual activity of the customer that owns the card.
Example: If a fraudster makes several large payments within the space of a few minutes, this is an abnormality that should be investigated more closely because it is not typical behavior for an ordinary customer.
The account will generally be frozen and the customer contacted to confirm it was them making the transactions. Several security questions will be asked to approve the customer’s identity and then the account freeze will either be lifted if it is found that the activity was that of the customer, or investigated if not.
The way that you use velocity checks depends on your circumstances. Many different types of date can be tracked, and you need to think about what are the most relevant to you based on your customer’s behaviour. Some examples may include.
- How much has been spent by a customer in the last day?
- How many times has a transaction been tracked in the last day?
- From one device, how many transactions have been tracked in the last day?
- How many transactions have come from the same IP address in the last day?
These examples all have a tracking identifier, a number and then a time frame to track against. Normal allowances can then be added to the system, and any activity that falls out of this ‘normal’ range will be flagged to be reviewed.
Velocity checks are very useful but not the only solution to fraud. Make sure to also look into other options like managing chargebacks and what to do after transaction fraud is detected.
