What is Transaction Monitoring in AML?

By Colin Bristow, Principal Pre/Post-Sales Systems Engineer, CFS Pre Sales Support

What is transaction monitoring in AML (Anti-money laundering)?

Transaction monitoring refers to the monitoring of customer transactions, including assessing historical/current customer information and interactions to provide a complete picture of customer activity. This can include transfers, deposits, and withdrawals. Most financial firms will use software to automatically analyse this data.

Potentially, the most effective approach to transaction monitoring in AML would be to have an employee manually stop and interrogate every transaction completed by a customer. Only after this review would the transaction be authorised for completion.

While the sheer amount of resources required makes this a ridiculous proposition, some organisations could be exposing themselves to potentially greater levels of risk through the automated approach adopted for their transaction monitoring systems (TMS).

The role of transaction monitoring systems in AML

A TMS of some description has been a mainstay process within firms for many years, providing risk-based AML transaction monitoring. The TMS will typically use information from know your customer (KYC) processes to account for the client risk. The risk measures are then used as part of rules/scenarios to identify certain account-based activities for investigation and possible disclosure.

More often than not, the approach to a TMS is only reviewed following some form of sanction/investigation by the regulators.

As a result of infrequent TMS review, three issues arise:

False Positives: the number of cases highlighted by the TMS that do not warrant review.

One of the main areas of operational cost relates to the number of people who need to review the output from the TMS. If the system is creating a large number of cases which do not warrant review; the operational costs increase, potentially creating a backlog and risking legitimate cases not being investigated in a timely manner.

Depending on the approach to TMS, true detection rates of cases worthy of investigation can be between 0.5 to 7%. Over 80% of those were submitted by credit institutions - banks. Using the detection rates outlined above, this means firms as a whole could be investigating between 8.8 - 124 million false cases.

Addressing false positives through regular review of output, false positive ratios and also the quality of scenarios, is a major part of effectively managing transaction monitoring systems in AML. Detailed usage of analytics is typically the best approach to managing this.

One size fits all: clients and activities grouped with the application of a single scenario that applies to all.

The impact of this approach is typically an increase in false positives over time; it also indicates a gap in monitoring. Even if clients are segmented into similar accounts or business types, there are normally lower levels of granularity associated with the segmentation. Instead, firms can adopt a more analytical approach to dynamic segmentation. This allows firms to use underlying characteristics of the client activity (e.g. transactional behaviour) to describe more focused segments - and then scenario/thresholds settings.

Updates to the segments may only occur once every 12-18 months., Being able to describe the segments in greater detail will provide greater clarity. Moreover, it also provides the opportunity to apply greater targeting of scenarios/rules to be applied.

Too many rules/scenarios: too many must be contextualised on the firm, but a growing number of scenarios can mean duplication of efforts.

As changes in the internal and external business environment occur, there can be a tendency to increase the number of scenarios. Over a period, the number  of scenarios has increases. This presents a twofold issue: Firstly, management/administration of the scenarios becomes challenging in terms of understanding which to review and when. Secondly, the overlap between scenarios and potential duplication of case creation becomes an

The management/administration of scenarios is key to understanding the performance of each - and provides an indication of effectiveness. If there are too many scenarios, a risk is that timing the for review of scenarios can become ad-hoc and poorly directed.

Overlap between scenarios, typically duplicating cases for investigation, also becomes a significant issue. If a customer has different accounts which create cases, centralising the cases into a single customer view (if not already catered for the in TMS approach) becomes an overhead. In addition, understanding where and by how much scenarios overlap can provide additional clarity relating to gaps in the compliance coverage and where adjustments can be made to scenarios to cater for the overlap.

How SAS' approach to transaction monitoring in AML can help your business

Application of advanced analytics to improving TMS processes will deliver benefits. As a leader in analytics, SAS is well placed to offer technologies and process to support an ongoing process review.. With SAS, you can increase the coverage of customer transaction activity while reducing false positive alerts while managing the risk of regulatory penalties. 

Free white paper

How AI and Machine Learning Are Redefining Anti-Money Laundering The bad actors are making strong plays. Take every advantage to fight back.
Download the paper

Read More

  • Find out how analytics can be used for AML compliance programs
  • Find out more about how SAS can support your business to improve the existing approach to anti-money laundering transaction monitoring.