Elevate on Deploying Behavioral Analytics as Part of a Multilayered Security Stack

Fraud is a constant challenge for organizations of all types, especially those that handle clients’ money or personal information. Either one can be a gold mine for bad actors leveraging account takeovers, new account fraud, phishing, social engineering or a wide variety of tactics to break into company systems. 

One company that is conversant with the fraud threat is Elevate, a platform for lenders based out of Fort Worth, Texas. The company has had its run-ins with fraudsters in the past, but it was able to defeat them successfully with help from a multilayered fraud prevention system featuring behavioral analytics. 

“There was an organized fraud ring that attacked our system last year, and although we caught them fairly quickly, they got through a lot of fraud checks,” said Ryan Prince, senior data scientist at Elevate. “They got through many layers of verifications that we did at the time, so that’s [when] we enhanced our tools even further and began to work with Neuro-ID.” 

Prince offered PYMNTS an inside look at the daily fraud threats facing businesses such as Elevate and why multitiered prevention systems using behavioral analytics are best-suited for keeping bad actors at bay. 

Fraud threats and weaknesses in legacy systems 

Fraudsters have been leveraging a growing array of tactics to break into systems at companies such as Elevate. The most dangerous of these techniques are organized hit-and-run attacks that attempt to saturate automated defenses with multiple fake applications at once. 

“Organized groups get stolen information, and then, out of nowhere, we’ll get eight or 10 applications within a couple of hours that all share similar characteristics or journeys,” said Prince. “The idea for them is that they get as much money as possible as quickly as possible.” 

Single-layer security systems can easily be overwhelmed by these types of attacks, as they lack the capacity to be everywhere at once. A multilayered approach is far superior to ensure that any false applications that slip through are caught by a second or third line of defense. 

“More checks are obviously better from a fraud perspective, especially more checks that don’t add friction,” said Prince. “Most people are familiar with [multifactor] authentication, where we send a text to verify yourself, but it would be even better to have additional checks that don’t add any friction [for] the customer. Just analyzing the behavior throughout the application is so helpful because we’re able to cut out that fraud without [introducing any] additional [hurdles on] the customer side.” 

Behavioral analytics is Elevate’s most valuable tool for providing that extra, frictionless authentication layer. 

How behavioral analytics fits in

Elevate’s behavioral analytics system inspects user data as it is typed in, including names, usernames, passwords and email addresses. It provides the extra authentication layer by examining exactly how these are entered, with irregular spellings, typos or hesitancy signaling that fraud could be at work.

“The organized fraudsters are fairly automated, writing a macro to scrape all the information from a [text file] and import it into an application,” Prince explained. “Having something that actually looks at the behavior and the characteristics of somebody applying helps add a certain level of security there and tries to analyze: Is there a real person doing this stuff, or is it happening too quickly? Or does it seem like these 12 applications did the exact same thing, which would be more characteristic of a computer program versus individuals?” 

Elevate sees a bright future for behavioral analytics in this arena, with even more features in development to drill deeper into fraud prevention. These systems can weed out fraudsters and help marketing teams determine user intent and reduce potential friction points in onboarding forms. 

“My biggest hopes are that we can essentially predict and identify the intent of a particular applicant based on how they work through the website,” said Prince. “In addition to that, I think there’s a marketing angle that helps us optimize the flow of the application to reduce the friction for good applicants as well.” 

As with security systems, behavioral analytics for marketing purposes would likely work best in tandem with other customer-facing applications. Redundancy is a critical part of any sophisticated system, and behavioral analytics ensures that nothing slips through the cracks.