During our first session of Fraud Office Hours, an attendee asked, "How can you detect anomalies in a sequence of payment transactions?" Watch this video clip to see how Unit21's Head of Fraud Risk, Alex Faivusovich, responded.
How to Detect Anomalies in a Sequence of Transactions
"Finding anomalies in transaction data is really the classic way to detect fraud.
And I think if we look at legacy solutions, what used to be on the market fifteen years ago, you wouldn't have been able to profile a user. You won't be able to understand how your user behaves in certain situations.
But then, let's say ten years ago, solutions were finally able to allow you to create some types of profiling. So you were able to take, let's say, certain payment methods or certain MCC codes or certain activities and profile those for the specific user and then write rules based on the profiled user.
However, what we're doing today with Unit21 is basically that we allow our customers to run rules with pretty much any look-back period on your historical data that you want. And by doing that, you no longer need to manually create those user profiles because we already have your historical data. The way our rules engine works is that it knows how to go back and look at the activity, the specific activity, that you try to detect and understand how this user had been interacting so far.
So, for example, if you try to flesh out maybe, let's say, a customer who suddenly starts depositing too much cash, injecting cash into the system. It's very easy to use Unit21 to understand how this person was doing this in the past and can set up a trigger to flag this type of anomalous behavior for review."
Interested in seeing how this works first-hand? Check out our first session of Fraud Office Hours on-demand for a quick demo of Unit21's software: