Compliance with Know Your Customer (KYC) and Anti-Money Laundering (AML) laws and regulations have necessitated the evolution of transaction monitoring systems. A financial institution must have a foolproof and adaptable process for monitoring customers’ transactions and identifying suspicious behaviors to mitigate financial crime risks, prevent fraudulent activities and guarantee full compliance with banking regulations — particularly in the face of rapid and expanding digitization of financial transactions.
The biggest challenge when fighting financial crimes through transaction monitoring (TM) is that criminals always find ways to adapt and go around existing transaction monitoring solutions. For the timely prevention or detection of financial crimes, a financial institution must be one step ahead of criminals and not the other way around.
What are the pitfalls of current transaction monitoring solutions, and how can they be remedied?
4 Pitfalls of Current Transaction Monitoring Systems
Implementation and maintenance of transaction monitoring systems have been the standard operating procedure within many organizations for quite a long time now. The International Compliance Association (ICA) reports that “firms use a blend of manual and automated systems, depending on the volume and complexity of the transactions they process, their risk appetite, and their level of maturity. The value of the financial investments allocated to transaction monitoring’s technical and human resources is a key factor informing the quality of the systems and controls and the relative level of sophistication.”
The aforementioned “quality of the systems and controls and the relative level of sophistication” are not always good enough, however, to ensure the effectiveness of the transaction monitoring processes that have been put in place.
Below are some common pitfalls of current transaction monitoring systems:
1. Limited Customization Capabilities
Most transaction monitoring systems currently being used by institutions are off-the-shelf, generic software. Unfortunately, these types of software are designed to address common risks and, therefore, often prove wasteful and ineffective. This is because they do not take into account risk exposures that are unique to an organization and its clients as well as new risks that have evolved with the rapid digitization of financial transactions.
2. Preset Rules That Deliver False Positives
Preset rules are part of generic transaction monitoring systems. They are used to process a high volume of client data to identify specific transactions that appear suspicious, such as when monitoring bank transactions. Rule thresholds can be customized based on an institution’s known potential risks. However, they are still prone to inaccuracies — producing too many false positives — because the rules incorrectly assume that client profiles and behaviors are always consistent. Fixing a broken system and sorting through these false positives can be costly in terms of time, staffing and money.
3. Ineffective Use of AI or Machine Learning in Transaction Monitoring Systems
AI solves the problem of inaccuracy inherent in traditional transaction monitoring solutions designed with preset rules. However, poor handling of data, which the AI will rely on, renders the AI ineffective.
4. Transaction Monitoring Regulatory Requirements
What is considered acceptable behavior by one regulator from one jurisdiction may be considered unacceptable and necessitate a review by another regulator from another jurisdiction. These different regulatory approaches can become a problem for institutions that operate across different jurisdictions when monitoring bank transactions and other financial activities.
The Changing Landscape of Transaction Monitoring
Most transaction monitoring systems still use a static approach based on predefined rules designed to detect suspicious behaviors, e.g., disguising funds, illicit fund transfers, payment structuring, etc. With systems that allow customization of rules, a financial institution may select the rules that are best suited to their needs and business risks.
For example, an institution can opt for a top-down approach. This approach involves the configuration of rules for specific client categories, each of which is based on known attributes. With the top-down approach, a customer that is considered high-risk will be subjected to a high level of scrutiny.
On the other hand, the bottom-up approach is based on data mining and unsupervised modeling. Transactional behaviors are used to categorize clients into clusters, and a data-led tuning process is used to set thresholds to maximize true positives and minimize false positives.
“An effective transaction monitoring system may employ an automated approach but will still rely on a certain level of manual, human intervention to review potential issues identified.” This was one of the key considerations for a transaction monitoring system recommended by ICA.
As electronic financial transactions are becoming increasingly more convenient and accessible, criminals are also finding new and complex ways to commit financial fraud — rendering a static transaction monitoring system even more ineffective. Therefore, switching to a dynamic and automated system has become an urgent necessity.
Dynamic Transaction Monitoring
Even with the option to customize rule thresholds, static transaction monitoring systems are still limited by preset rules. Meanwhile, criminal behaviors evolve, especially in response to technologies that are designed to detect them. And with the volumes of cashless and online transactions growing year in and year out, there has also been a considerable increase in the number of suspicious transactions that require review. This means that transaction monitoring solutions must also be adaptable and dynamic to remain effective in the face of rapidly evolving trends in the financial sector.
Transaction monitoring based on machine learning and AI is the best option to give financial institutions better AML compliance and leverage. It’s essential to remember that these transaction moniroting solutions still require a sufficient understanding of customer behavior and AML risks if dynamic transaction monitoring is to be effectively implemented.
According to a Deloitte article on the dynamic approach to transaction monitoring:
“The solutions themselves are typically not built with the functionality to consider real-world experiences that are growing and changing dynamically. For an AML turnkey solution to be utilized dynamically, detailed subject matter knowledge must be leveraged to enhance the true potential of the solution. By coupling real-world experiences with powerful, automated software, the transaction monitoring process will be transformed to keep pace with ever-evolving financial crime activity.”
In short, an effectively dynamic and machine learning- or AI-based transaction monitoring solution must engage in continuous transaction monitoring.
What Is Continuous Transaction Monitoring?
Continuous transaction monitoring or continuous monitoring is the automated process of continuously monitoring financial transactions and internal controls that are put in place to detect existing and potential risks. Continuous monitoring facilitates the quick and timely identification of anomalous business processes or individual transactions so they can be given the appropriate attention and resources for effective resolution.
Continuous monitoring relies on predictive machine learning and AI technology, advanced analytics and key transaction information, which is automatically extracted and accumulated from enterprise resource planning (ERP) systems across an organization. Through continuous monitoring, huge volumes of transactional data are consistently analyzed across disparate systems, such as devices, and over different time horizons. As a result, potentially troublesome patterns are detected so they can be promptly and proactively addressed before they develop into a more serious problem.
Continuous monitoring makes it possible for financial institutions to more quickly and effectively adapt and respond to the ever-changing risk and regulatory climate. The greater visibility into business processes and financial transactions afforded by continuous monitoring also translates into greater transparency for stakeholders, investors and directors. As a result, continuous monitoring encourages bigger investments, promotes positive publicity and improves overall compliance.
Instnt offers continuous identity assurance in the form of Instnt Verify™, a step above and beyond continuous transaction monitoring. Instnt Verify™ is a more secure method that enables businesses to verify transactions from authorized account owners. The Instnt Verify™ solution prevents account takeovers and aids anti-money laundering efforts. This crucial tool, only available at Instnt, helps businesses maintain identity assurance starting from account opening and continuing through all transactions going forward.
Equip Your Business for the Future With Instnt
Instnt Verify™ is one of Instnt’s many powerful advantages that you can equip your business with to optimize your transaction and identity monitoring for the ever-changing fraud climate. Along with the suite of Instnt fraud prevention and compliance solutions, it helps deliver 40% fewer false positives. Stay one step ahead of potential threats by signing up for a free demo today!