BioCatch Examines Behavioral Predictors of New Account Fraud

BioCatch has released a short primer that looks at some of the predictive factors for new account fraud. The post specifically explains how behavioral analytics can offer a glimpse at a user’s short and long-term memory, which in turn can help determine whether or not that user is a legitimate customer or a fraudster.

In that regard, behavioral biometrics bases its conclusions on neuroscience and the cognitive tendencies of various users. For example, most people are quite familiar with their own personal information, and do not need to look up basic details like their address. As a result, they will move at a steady pace while filling in an account opening form.

Fraudsters, on the other hand, are typically using stolen information, and will seldom have it memorized. They will need to refer to source material to fill in each field, and that creates a segmented typing pattern as they jump from the registration form to a secondary reference like a spreadsheet. However, they will be more familiar with the registration form itself, and with computer shortcuts more generally, largely because they have been through the process more times than the average user. Fraudsters are also more likely to rely on copy-paste techniques rather than manual entry, for many of the same reasons.

According to BioCatch, segmented typing was flagged as a risk factor in 44 percent of all new account fraud cases, while a user with a segmented typing pattern was four times more likely to be a fraudster. That makes it an extremely useful metric for organizations that are trying to protect their customers and their account opening processes.

This is not the first time that BioCatch has warned about new account fraud in the past few months. The company has previously noted that social engineering attacks and mule accounts have both increased during the COVID-19 pandemic.

–

August 6, 2020 – by Eric Weiss