Transaction monitoring rules: navigating the future

Eglė Kontautaitė
Author
Eglė Kontautaitė
Published
February 26, 2024
Transaction monitoring rules

In the dynamic and ever-evolving financial sector, the emphasis on robust Anti-Money Laundering and Countering Terrorism Financing (AML/CTF) strategies has intensified, with particular emphasis on transaction monitoring rules and scenarios.

The cornerstone of these strategies is the strengthening of AML/CTF transaction monitoring solutions and developing more advanced suspicious transaction identification scenarios. AMLYZE, a standout in Software as a Service expertise, not only delivers automated transaction monitoring solutions to detect suspicious patterns indicating financial crime but also offers comprehensive guidance to financial institutions.

This includes assistance in implementing necessary rules and advanced scenarios tailored to reflect financial institution’s business model and associated risks herein. It is our mission to develop, implement, and continually refine effective AML/CTF monitoring systems to combat money laundering and financial crimes. We have gathered a deep expertise on developing sophisticated transaction monitoring rules and scenarios by cooperating with financial institutions, financial intelligence units and regulators.

The essence of AML/CTF transaction monitoring

Transaction monitoring is like the security guard of a financial institution’s system to prevent money laundering and terrorism financing. It keeps an eye on all the financial activities to spot any suspicious behavior and alert on it for further analysis. Ultimately, a human analyst reviews these alerts to decide if the information additionally provided by the customer indeed  supports the initial suspicion and the transaction should be reported to the financial intelligence unit as a possible criminal activity. Think of transaction monitoring like having a magical crystal ball that can predict if someone is trying to launder money. However all magic comes with a price. In this case the price is the meticulously crafted transaction monitoring rules and scenarios.

The institution’s effort to predict the possible wrongdoing starts with the sourcing and analysis of the data: customer’s profile, financial transactions, customer’s behavioral patterns when accessing financial services. Analysis of data might be automated and produce automatic alerts on suspicious or at least on unusual customer’s behavior. Detection of suspicious patterns and prediction if those patterns are associated with money laundering or even a predicate offense of money laundering (such as fraud, tax evasion, sanction evasion etc.) are initially based on automated transaction monitoring rules and scenarios. 

Risk-based transaction monitoring 

Setting up the “magic transaction monitoring ball” to function effectively is no easy feat. It involves creating intelligent rules and scenarios that require a deep understanding of criminal tactics (typologies used for money laundering) and strong data engineering skills. These rules are crafted to detect any unusual or suspicious patterns within the large volume of transactions being processed. When an abnormal pattern is detected, it triggers alerts for AML analysts to conduct further investigation, helping to uncover criminal activity and freeze suspicious assets before it’s too late.

Money laundering, terrorism financing, and other financial crime patterns are typically identified and shared by law enforcement, financial intelligence units, and regulators through various channels like supranational, national risk assessments,  reports and dedicated feedback to institutions. However, it’s up to institutions to interpret and implement these patterns into specific transaction monitoring rules. Regulatory bodies often encourage a risk-based approach, allowing institutions to enhance basic rules with more sophisticated scenarios aligned with risks to better detect and prevent illicit activities.

The risk-based approach integrates various risk concepts that might be applied differently within transaction monitoring processes. We are of the opinion that the risk-based approach begins with the segmentation of customers. Categorizing customers into different risk groups, allows the creation of tailored transaction monitoring rules that address the specific risks associated with each group. Customer segmentation is crucial for profiling customers, enabling the identification of unusual transactions or patterns. It’s important to compare a customer against similar ones to detect abnormal behavior accurately.

In the end, transaction monitoring revolves around assessing transaction risk. Segmenting customers is the first step in the risk-based approach, which can be further refined by employing various algorithms within the transaction monitoring rules to identify different aspects of transaction risk. Moreover, the transaction monitoring rules themselves may undergo risk scoring to prioritize the investigation of alerts based on different rule sets. 

Developing efficient monitoring rules

Looking for suspicious patterns can sometimes feel like searching for a needle in a haystack, and many AML analysts grapple with a high number of false positives caused by ineffective transaction monitoring rules.  In financial institutions, AML departments operate within certain resource constraints. Therefore effective transaction monitoring rules and scenarios are essential not only for successful fight against crime, but also for efficient allocation of human resources for further analysis of alerts. 

A transaction can appear unusual from various viewpoints, and transaction monitoring rules assess it from multiple angles. Certain rule algorithms analyze the account’s historical usage to identify any sudden deviations from established patterns, while others consider factors like transaction size and frequency within a specific timeframe. However, the challenge lies in determining whether these deviations are genuinely suspicious or simply reflect gradual fluctuations in transaction volume or seasonal variations. There are no one-size-fits-all solution, however we might provide several suggestions on how to make transaction monitoring rules work more effectively.

  1. Avoid on-size-fits-all approach, use customer risk segmentation. Apply for each customer risk segment specific rule set.
  2. Utilize data analytics for rule development. Comprehensive data analysis is essential for understanding patterns and trends within the institution’s client portfolio as well as individual client behavior.
  3. Utilize easily adaptable rule parameters within the system. These parameters should be flexible and capable of adjusting to evolving compliance requirements. Avoid rigid, hard-coded parameters, as parameterization should facilitate swift updates in response to shifts in risk profiles or system performance.
  4. Opt for dynamic rules over static fixed parameter rules to enhance effectiveness. Static thresholds (such as “create an alert when the amount exceeds a <fixed threshold>”) often lead to numerous false positives due to their rigid nature. Conversely, rules with automatic threshold adjustments are better equipped to detect unusual behavior with greater accuracy. These dynamic rules vary in complexity and yield fewer false positives. For instance, a rule like “alert when the amount exceeds two times the standard deviation for <specific risk segment> in the last x days” offers more nuanced detection capabilities.
  5. Merge rules into rule sets or scenarios to more accurately capture specific criminal typologies. While individual rules may not fully capture certain money laundering typologies on their own, combining them within a rule set or scenario can enhance detection capabilities. This approach allows for a more comprehensive analysis of transaction data and improves the identification of complex money laundering schemes.
  6. Carefully choose the rules for real-time transaction monitoring to swiftly identify and freeze transactions with a high risk of money laundering. However, freezing transactions without valid grounds could burden clients accessing financial services unnecessarily. It’s essential to consider which transaction monitoring rules will trigger the freezing of transactions or funds, and which will serve as indicators for retrospective client behavior review without immediate asset freezing.
  7. Backtesting isn’t just a buzzword or a regulatory checkbox; it’s a valuable tool. It’s used to verify if transaction monitoring rules yield expected results based on historical data.

These are just a few basic tips and tricks. However, in an ideal scenario, artificial intelligence (AI) would function as an indispensable tool for detecting complex transaction patterns, promptly alerting to clear criminal typologies, and establishing an AML framework devoid of false positives — a goal cherished by every monitoring specialist.as AI continues to advance across various domains, it’s becoming evident that these methods may not suffice to address evolving threats in the criminal world. To better tackle such challenges, more comprehensive algorithms incorporating machine learning are necessary for transaction monitoring rule and scenario development.

The new trends in enhancing monitoring

AI and machine learning have permeated every realm of automation, including transaction monitoring. In an ideal scenario, AI would serve as an indispensable tool for detecting intricate transaction patterns and promptly alerting on clear criminal typologies and creating an AML realm with no false positives (a dream of every monitoring specialist).

While the goal of replacing the human element in AML processes is an aspiration, it remains largely in the visionary stage. Machine learning, AI, forecasting models (to name a few) rely heavily on the data fed into the system and the data from decisions made. Access to high-quality, comprehensive datasets is paramount for effectively training AI models. These datasets should encompass various transaction types, customer profiles, historical patterns of suspicious activity, and decisions made by humans when alerted transactions were deemed suspicious or exhibited criminal patterns. In essence, it’s akin to the proverb “it takes a village to raise a child,” translated into AI terms as “it takes a vast pool of data to educate an AI.” In the realm of suspicions and transactions, a public-private partnership (involving financial intelligence units collaborating with financial institutions) is instrumental in producing such datasets. And AMLYZE is the one, who is taking seriously the possibilities of this public-private partnerships might bring (yes, this is a spoiler for our future articles).

AI has become a buzzword across various industries, but the reality is that many institutions are still grappling with enhancing their expertise in advanced analytics and implementing well-structured algorithms. Processes like robotization and backtesting automation, while seemingly straightforward, represent “new technological trends” for numerous institutions. Analytical software, such as that developed by SaaS, is already supporting analysts in data analysis and automating manual tasks.

The widespread adoption of instant payments into customers’ daily lives presents new technological challenges for an accelerated transaction monitoring processing. Implementing technologies that enable fast transaction processing for instant payments without compromising on the effectiveness of transaction monitoring checks will be critical in fostering the evolution of both the payment market and AML technologies.

Future directions with AMLYZE

Predictive analytics for proactive monitoring and behavioral pattern analysis for early detection are at the forefront of AMLYZE’s approach. Collaboration within the financial community for best practices and regulatory partnerships is essential for maintaining compliance excellence.

In an era where financial crime tactics are increasingly sophisticated, financial institutions must remain vigilant and adaptive. AMLYZE’s comprehensive approach to AML transaction monitoring rules not only aligns with regulatory demands but also fortifies against financial crimes. By integrating advanced technologies, promoting continuous learning, and fostering collaborative efforts, institutions can safeguard their operations and uphold the integrity of the global financial system.

Read more here about the basics of transaction monitoring and the key steps in the process.

About the author

Eglė Kontautaitė
Author
Eglė Kontautaitė
Eglė is Head of Customer Solutions at AMLYZE and has more than 13 years of experience in the supervision of financial market participants, most recently as Head of AML Department at the Central Bank. Also former country representative at MONEYVAL.

Related

AML in crypto
Blog

AML in the crypto space

Explanation of the importance of AML measures and guidance on how to achieve them in the fast-changing crypto world.
by AMLYZE
12 min read
AML for Fintechs
Blog

AML for Fintechs: a detailed guide

Explaining the working model of Fintechs, the main myths surrounding the sector and the role of AML/CFT in their day-to-day operations.
by Eglė Kontautaitė
13 min read

Empower your compliance

Let us know how we can help

    Fill in the form bellow to contact us



    Why request a demo?

    It doesn’t matter whether you are interested in a complete end-to-end AML/CFT solution or just a single module from our range. We can help.

    Experience up to a 62% reduction in false positives

    Benefit from a library of over 100 risk rules

    Complete investigations in 3x less time than manually

    Save up to 3 hours per STRs/SARs filing

    Access a library of over 200 pre-defined scenarios