Call centres, especially in the financial sector, can significantly reduce frauds by installing voice biometrics solutions that work quietly in the background, says Mark Lazar
Contact centres in the financial industry are vulnerable to “professional fraudsters”, whose systematic attacks account for the majority of total fraud calls. Some contact centres are responding with a new generation of voice biometric systems that silently detect known fraudster voices during incoming calls. The goal of these new systems is to increase fraud protection without disturbing customer experience.
Contact centre vulnerability to professional fraudsters
Contact centres in the financial industry have seen growing attacks by professional fraudsters to breach accounts and conduct fraud. Traditional security questions don’t stop professional fraudsters, who are skilled in social engineering and armed with stolen identity data. Industry analysts now identify the contact centre as a security weak link exploited by fraudsters in the face of tightened online security.
Behind the professional fraudster’s success is the weakness of the security question paradigm. Today’s fraudsters buy stolen identities and leverage social networks to answer most security and out-of-wallet questions. Fraudsters also socially engineer agents, who have to play the conflicting roles of customer service (making callers happy) and security (being suspicious of callers). By repeatedly and systematically calling, fraudsters learn enough to pressure agents into security breaches. In fact, studies at various contact centres find that 70%-95% of total fraud calls are “repeat attacks” by a small set of professional fraudsters. Contact centre fraud is a classic 80/20 problem – most fraud calls are by a minority of the bad callers.
New voice biometric solution for fraudster detection
In response, contact centres at banks and card issuers are turning to voice biometric systems for fraud detection. Voice biometric technology works by profiling and recognising a human’s unique vocal characteristics, or “voice print”. [Human voices have mathematical representations that are unique and not easily masked without detection.] To perform fraud detection at contact centres, voice biometric systems store the voice prints of known fraudsters in a database, and then screen incoming calls against this database to detect a fraudster’s voice.
Upon detection, the system notifies the live agent or fraud management system. The key to this new generation of voice biometrics is its “passive” operation – calls are silently screened without caller interruption. No questions are asked about callers. Agents proceed normally until they are alerted to any fraudster detection by the system. Today’s passive operation of voice biometrics is a key advantage over historical systems, which were required to operate “actively” – interrupting callers with specific questions or instructions in order to screen calls.
Moreover, for higher performance, these systems combine voice biometrics with predictive analytics that factor in other non-biometric variables, including account or call meta-data With these advanced techniques, these systems can achieve accuracy at scale at large call volumes.
Benefits to the contact centre
Such voice biometric systems can provide significant benefits to the contact centre:
1. Stronger fraud protection – These systems can significantly reduce fraud dollar losses and the number of fraud attacks. That’s because voice-print screening detects “repeat” calls by known voices, and repeat calls constitute the majority of fraud attacks on the call centre. Therefore, the system is able to subvert the majority of these attacks, which lowers losses due to fraudulent calls and ultimately discourages those fraudsters, who move on to other banks. The “80/20” rule of contact centre fraud – the majority of fraud is frequent repeat calls by a minority of fraudsters – makes this a rewarding application for voice print tracking.
2. Greater fraud visibility and analytics – Voice biometric systems have the capability to track an individual fraudster’s voice even as it calls across multiple accounts and time. With that tracking capability, the system can become a rich database of fraudster activity and behaviour. The contact centre can gain visibility into fraud patterns or fraudster details that was not possible before. This may translate into insights that can improve contact centre security or efficiency.
3. Reduced vulnerability to social engineering – Although fraudsters can trick or coerce agents, their smooth talk doesn’t fool a voice biometric engine. By alerting agents during a call, or by alerting the fraud management system even after an agent has been fooled, the voice biometric system reduces the contact centre’s vulnerability to social engineering of agents.
As fraudsters have evolved with new technologies and techniques, so should the contact centre evolve its security technology. Call screening against a fraudster voice print database may be a quiet way to augment contact centre security without disturbing customers.
Mark Lazar is Global Vice President, Enterprise Risk Management, Verint Systems.
