fraud prevention in ecommerce

How to identify fraud and enhance security measures with behavioral analytics



Barely two months into 2024 security researchers have already identified what is being reported as “the mother of all breaches,” with more than 26 million records hacked worldwide from major brands and entities.  Bad actors are growing increasingly more sophisticated and bold in their quest to commit cyber-attacks. The repercussions are devastating, with the cost of a data breach worldwide averaging US$4.32 million in 2023 – and more than double that for the healthcare sector. Regardless of industry, understanding how to best spot anomalies that could indicate identity fraud is more important than ever. This is where behavioral analytics comes in.

Article at a glance:

  • What are behavioral analytics?
  • How are they being leveraged as a powerful tool to identify unusual patterns, detect potential fraud and enhance security measures?

Behavioral biometrics

What are behavioral analytics?

Behavioral analytics is a process that examines patterns of behavior and provides organizations insight into how their customers behave. To interpret and analyze behavior, data is collected from various sources, with profiles built based on known user behavior to determine if a user is acting like a normal consumer or to detect abnormal, potentially fraudulent behavior.  As customers engage online and interact with apps and websites, a pattern builds; anything and everything from the time of day a customer might log on, to how they move their mouse and type on a keyboard (i.e., behavioral biometrics) tells a story about the user. By employing machine learning algorithms, organizations can use this historical data collected to create predictive models, taking behavioral analytics one step further by predicting the likelihood of certain behaviors.

How is behavioral analysis used in fraud detection?

One of the most crucial roles that behavioral analytics plays in fraud detection is through user profiling, which involves the creation of profiles of normal user behavior based on historical data. Deviations from these established patterns and baseline behavior can then be automatically flagged for further investigation. Better still, this monitoring is done in real-time, enabling the immediate detection of suspicious activity and, in turn, fraud prevention. Behavioral analysis provides an extra layer of security, enabling organizations to more confidently identify a legitimate consumer versus an imposter.

How does Mastercard Identity enhance security measures with behavioral analytics?

Mastercard’s Identity Engine provides a comprehensive view of a customer’s digital identity. Comprised of two exclusive datasets, the Identity Graph and the Identity Network, the Engine examines the relationship and the validity of a customer’s identity elements and looks for behavioral patterns and anomalies, seeking to identify suspicious behavior and identity theft. Some questions that the Network seeks answers to, using behavioral analysis, include:

  • When was this email first seen in a digital interaction?
  • How risky is this IP address?
  • Has there been an increase in the number of interactions from this IP address (or email) in the last hour?
  • What other anomalies are present in the usage of these identity elements (such as name, email, phone, address and IP)?
online security and digital identity verification

Next, a risk score is calculated, in real-time, combining authoritative data such as match statuses, metadata and linkages from the Identity Network and the behavioral analysis of identity elements in the Identity Graph. With a scale of 0-500, with a higher number indicating a riskier transaction and a lower number indicating less risk, the risk score enables organizations to calculate a more informed and accurate risk assessment on each customer. Furthermore, an additional layer to Mastercard Identity’s solution is provided through device and user behavior analysis. This behavioral analytics tool examines the current and historical interactions of the customer and can be used to differentiate between a legitimate cardholder and an imposter. For example, the technology can analyze:

  • The volume of traffic coming from a single device
  • The method of moving through a form online
  • The pattern and speed of typing and whether it’s consistent with a human or bot
  • If a form field is autocompleted or copy and paste is used

To better understand how Mastercard Identity can help your organization leverage behavioral analytics to enhance security alongside KYC and 3DS, prevent fraud and improve overall decision making by better understanding patterns of human behavior, get in touch today.

Louisa Farrar Avatar

About the Author

Related content