How to Leverage Non-Monetary Indicators of Fraud
"It’s important to remember that every customer should display unique behavior; people don’t behave identically - and they shouldn’t.
In the last year, we’ve been very busy helping customers bring the idea of monitoring beyond simply tracking money movements. Instead, we’ve tried to help teams gain insights from data you have that isn’t related to money movement, and then integrate that into your fraud prevention strategy.
For some cases or customers, it could be something as complex as behavior analytics and heat maps, which lets you see how your customers utilize your app; where do they press, how do they fill in information, do they use copy-paste?
For some customers and in some use cases, it can be as simple as tracking the events that are happening on your platform and associating those events with money movements.
By looking at non-monetary indicators, we should be able to create a more sophisticated strategy.
If we can see that a user is doing something on the app or website that they don’t usually do, or we detect anomalies in their behavior, we can use that information to gain insights. By following that information, we can then see money movements that may not seem correlated to the way the user has interacted historically.
This enables risk and compliance teams to create more sophisticated strategies that better detect fraud.
When making this analysis, it’s important to remember that every customer should display unique behavior; people don’t behave identically - and they shouldn’t. For each customer, the onboarding journey is unique, they use products and features in a specific way, and they have unique ways of accessing your product and features.
Likely, they have a unique device and access from a unique network, etc. In fact, one of the challenges fraudsters face is actually creating this uniqueness to skirt detection systems.
Think about it; if a fraudster is using 100 to 200 different identities across a variety of platforms, it’s still very hard for them to be unique each and every time and to behave in a unique way, to use unique devices, and to use unique network access.
For fraud prevention teams, it’s important to have as much information at your fingertips to help when making decisions; the more signals you have, the more insights you can draw (and the more valuable and accurate these insights are)."
How Unit21 Helps Risk & Compliance Teams Leverage Non-Monetary Fraud Indicators
Building off our very popular network analysis functionality, Unit21 has developed a graph-based rules feature that empowers teams to analyze user behavior through the connections between users and their unique details. This helps teams flag customers where users share Personal Identifiable Information (PII).
Fraudsters that manage to steal synthetic IDs can be very difficult to trace, but they may still be using similar devices or IP addresses. Unit21 can help risk teams start alerting on customers even if they haven't done anything transaction-wise—simply based on the fact that we see a few entities or accounts sharing the same device or IP address.
Teams can adjust thresholds of how many entities need to be tied to that same device or IP address to refine their search. We can then help teams find patterns where multiple customers are exhibiting similar unique qualities, such as the device, IP address, socials, name, and even geolocations. Unit21 helps teams detect fraudulent accounts before they are extracting money from the business.
For example, three or four users sharing the same address could make sense for a family, but 10 to 15 users at the same address would be far less likely. Refining rules can allow teams to better detect potential fraud using non-monetary indicators based on available PII, allowing teams to take action before fraud actually occurs.
Looking for more insights? Check out our fourth session of Fraud Office Hours on-demand for a deeper dive into how to use Unit21's platform to detect potential fraud using non-monetary indicators.
Subscribe to our Blog!
Please fill out the form below: