Proactive Fraud Detection: The Whole is Greater Than the Sum of Its Parts

In late April, NICE convened Interactions 2019 and all roads to secure and convenient customer emanated from three foundational elements: "The Cloud, Analytics, 'Digital & AI' and Automation." For more detail on the event see this post. Of these major components, the speech analytics resources developed by NICE Nexidia, which was acquired three years ago, were of special interest to me, especially as I home into how analytics is being elegantly integrated with the company's internally-developed, market leading voice-based identification and authentication capabilities.

As stand-alone products, the two offerings have proven their value for NICE's customer base. Historically, NICE Nexidia's core technologies had become invaluable to contact center managers to identify points in conversations when an agent goes off script or a customer says the keywords that indicate dissatisfaction or the intention to cancel a service. Meanwhile, NICE Real Time Authentication and Fraud Detection applies a totally different analytical engine (voice biometrics) to identify unique traits in a caller's spoken words and speaking patterns that betray them to be bad actors.

The two sets of product offerings share one thing in common: They detect patterns, especially anomalies, in the interactions between agents and callers that require attention and remediation.

"We've Already Authenticated You with Your Voice"

The big stage at Interaction 2019 was a giant demo space for NICE's contact center offerings. One of the most impressive use cases involved a parent booking travel arrangements for a family trip. Much of the demo involved the use of "conversational AI" to prompt a live agent to ask all the right questions, but I was most impressed when the conversation was propelled toward a purchase involving a live agent.

At that point the customer was transferred to an agent in order to complete a transaction. The IVR system said, "We've already authenticated you with your voice. You may now proceed with your transaction."

It was a clever callout of the effortless authentication made possible by NICE's Real Time Authentication (RTA) solution. While this showcased the "real time authentication" abilities of the platform, it also served to raise awareness of the power of voice-based authentication to prevent fraud. In front 3,000+ attendees, NICE made a point of showing how easily voice biometrics can be introduced into both the customer and the agent's respective workflows.

There were also breakout sessions featuring real-world implementations of RTA. Presentations by executives from Optum, the fast-growing information and technology subsidiary of UnitedHealth Group; M&T Bank; and Blue Cross/Blue Shield of Michigan explained the why passive, voice-based authentication and fraud prevention are so attractive today.

M&T Bank, one of the "Top 50" banks in the U.S. by several measures, described its implementation of voice biometrics for fraud detection, including how it enjoyed immediate, positive results. NICE's technology identified a repeated fraudster trying to gain access on day one of the deployment and was able to avoid losses associated with his or her access.

M&T, using the fraud detection technology from NICE RTA, took advantage of the fact that fraudsters are creatures of habit. A small cadre of them repeatedly call a bank's contact center with the intent of doing financial harm. For this reason, a database of voiceprints for a "black list" of fraudsters is a powerful tool for financial institutions.

NICE's proactive approach to fraud detection enables companies to expose unknown fraudsters hiding in historical recordings and automatically add them to the black list.  This ongoing response to sophisticated fraud attacks, the creation of high-quality fraudster watchlists from historical recordings, provides a highly effective defense.

The marriage of voice biometrics with NICE Nexidia's speech analytics is of particular interest because it can help make fraud prevention more predictive. It combines the fact that each of us have unique attributes to our speech patterns and spoken words (voice biometrics) with the fact that the words that an individual uses and how they say them are both consistent and predictable.

Thus the combination of NICE Nexidia and NICE voice biometrics heralds the day of "Proactive Fraudster Exposure". This approach serves as another example of how a combination of existing, proven technologies comprise a single solution can deliver results that exceed the capabilities of any of them operating separately. Offering an enterprise-grade portfolio solution propels fraud prevention to another level where an enterprise can be proactively more profitable in its approach to fraud-loss prevention and protection of customer data.