Automated Financial Crime Technology can fix the bottleneck of AML transaction monitoring
Risk and compliance teams are not only fighting to stay ahead of financial criminals. Financieele Dagblad kicked off the new year with an article highlighting the struggle facing Dutch banks to recruit money laundering experts. According to the article, banks are facing challenges not only to fill these positions but also to retain staff.
Barclay Simpson, an international recruitment company specializing in risk and compliance, published a report in 2019 that looks at current trends in the compliance market workforce. From their global research of 1,600 professionals, 44% stated that their departments did not feel adequately resourced. This statistic is across the industry and also includes Financial Crime Compliance operations and investigators.
Today, banks still recruit thousands of people to manually review the millions of transactions flagged every year as potential money laundering attempts. Simultaneously, banks are also struggling to deal with a different problem – inefficient Anti-Money Laundering (AML) systems that produce a flood of suspicious activity alerts containing large numbers of false positives.
Keeping up with regulations when skills are in short supply
Due to the nature of the financial services industry, there is a high requirement for continuous updates and amendments to financial regulations that protect customers and businesses from financial crime. And regulatory compliance is mandatory for all financial institutions. This creates additional pressure on teams that already lack the necessary number of experts to manage the increasing workload.
Meanwhile, criminals are constantly working to expose weaknesses in financial systems in order to launder their illicit gains.
Attracting and retaining talent
Added to this pressure the Barclay Simpson report also shows that identifying the right technical skills is the biggest recruitment challenge for 59% of employers surveyed. Attracting and retaining talent for these positions is tough when demand is high. To succeed at staff retention, financial institutions need to start by finding ways to make the work more fulfilling and combat fatigue caused by sifting through thousands of false positive alerts. Technology becomes very useful in this regard as it can help by automating and supporting repetitive tasks, so that investigators can add maximum value by focusing on the most critical cases.
Despite the recruitment challenges, there are many ways in which financial institutions can make the most of current operations to be efficient and effective at fighting financial crime, including money laundering.
The needle in a haystack
The banking industry has historically relied on pre-set rules to identify potentially suspicious activity. Rules-based AML systems cast a very wide net. Yes, they detect some financial crime, but they also produce a huge volume of false positive alerts. These are the alerts that, once investigated, turn out to be legitimate transactions but somehow breached one or other of the bank’s rules. Without the benefit of machine learning technologies, a large percentage of alerts generated could be false positives, yet they still need to be investigated and discounted.
The sheer volume of false positive alerts makes finding and focusing on real suspicious activity that much harder. The first step towards more effective money laundering detection should be for financial institutions to integrate machine learning-based adaptive behavioral analytics alongside traditional rules-based methods. It is an easy win, immediately eliminating a high percentage of false positive alerts and making the elusive 'needle' much easier to find.
Intelligent, self-learning technologies can speed up detection
A good place to start eliminating bottlenecks is looking for ways to adopt best-in-class technology that supports the work of financial crime investigators. ARIC™ Risk Hub is a machine learning solution that uses Adaptive Behavioral Analytics to focus on monitoring 'good' behavior and detect anomalies as they happen. The intelligent, self-learning technology increases the quality and accuracy of alerts raised because suspicious activity immediately stands out. This means the most urgent alerts can be flagged accurately and prioritized, reducing the workload and enabling teams to focus on the most important alerts.
Expert human judgment should be the key method used to decide whether suspicious behavior must be filed as a Suspicious Activity Report (SAR). Add this human expertise with intelligent machine learning models and rules, and the improvements will be instantly evident. This presents a compelling case for financial institutions to be more effective and efficient, underpinned by a scalable, future-proof risk management process.
The technology already exists to support these advancements. For teams, it offers a means to have more meaningful work, providing a feeling of satisfaction and ultimately a sense of accomplishment. These factors go a long way towards maintaining employee satisfaction levels, and of course, retention rates.
Featurespace has been recognized as a Strong Performer in The Forrester Wave™: Anti-Money Laundering (AML), Q3 2019 report. Read more in your complimentary copy of the report.
Leveraging machine learning technologies to detect and prevent financial crime in real-time can have a tremendously positive impact on the success rate of your risk prevention strategy.
Aite Fraud and AML Machine Learning Platform Evaluation
Aite Group, a global research and advisory firm, has recognized Featurespace™ as Best-in-Class in its 2019 report on fraud and AML machine learning platform vendors.
About Featurespace – www.featurespace.com
Headquartered in the U.S. and U.K. and with offices in Atlanta, Cambridge and London, Featurespace™ is the world-leader in fraud prevention and creator of the ARIC™ platform, a real-time AI machine learning software that risk scores events in more than 180 countries.
The ARIC platform combines adaptive behavioural analytics and anomaly detection to automatically identify risk and catch new attacks as they happen. The increased accuracy of understanding behavior strikes the balance between improving fraud and risk detection and operational efficiencies, while also reducing the number of genuine transactions that would be incorrectly declined by as much as 70 percent.