Transaction Monitoring Optimisation

The fast and effective detection of financial crime is an important part of any transaction monitoring process nowadays. KPMG is breaking through and changing the future of transaction monitoring with new automated solutions.

The current challenge for banks is a combination of ever-expanding and evolving regulations and laws that create a complex regulatory environment in which to operate and comply. Ineffective compliance is therefore no longer an option.

The optimisation of transaction monitoring systems is a complex task that is compounded by increased expectations from regulators. Financial institutions often carry out regular reconciliations based on simplified tests. Inefficient systems lead to a multitude of alerts. Globally, banks manually screen millions of alerts per month with a potential connection to financial crime, with nearly 95% of the alerts triggered being "uneventful". The review process of current alerts often requires large teams of staff, often spread across different geographical locations. Many of the compliance experts surveyed expect a further increase in staff numbers this year in order to maintain the current level of compliance.

In light of increasing and changing regulation and guidelines, rising transaction volumes and the identification of new money laundering typologies, KPMG has developed solutions that target both system optimisation and hit verification processes. The aim is to increase the quality of monitoring system performance and to provide faster, more accurate and more consistent verification of hit notifications, helping financial institutions meet their transaction monitoring obligations with higher quality and less effort.

Continuous review and optimisation of transaction monitoring systems can result in the production of lower volumes of higher quality alerts that can be more easily verified, either by humans or by using the Transaction Monitoring Alert Classifier.

• Automated optimisation is based on the use of machine learning to run thousands of what-if scenarios to determine the optimal state for customer segmentation, scenarios, rules and thresholds.
• These recommendations can be reviewed by experienced users to determine whether adjustments should be made to the transaction monitoring system based on the client’s risk appetite.
• The automated optimisation tool can also be used to model business changes to determine impacts on current monitoring scenarios. This insight enables informed decisions about business changes, resource needs and managing a high number of hit messages.

Automated optimisation infographic

The Transaction Monitoring Alert Classifier automates the decision-making process for hit verification within the first line of defence.

• Using advanced machine learning methods, the tool automates the identification of alerts that are likely to require closer investigation (automatic escalation of critical messages) and alerts that are not suspicious (automatic closure), allowing analysts to focus on high risk activities.
• Every decision to review alerts has a confidence-security level and is supported by a human-readable decision basis. Clients can therefore tailor their use to specify coverage and accuracy rates that support their risk appetite.
• The use of monitored machine learning ensures that decisions are transparent and can be understood and reviewed by auditors and regulators.

Management information and analytics provide insights into system performance, data quality and additional intelligence that can be relevant to financial crime controls. 

• Software packages can be developed that provide insight into payment activities and highlight risk areas in transaction behaviour.
• Visually appealing, customisable dashboards present management information in a way that highlights key insights and enables data-driven decisions.

KPMG has combined expertise in financial services, transaction monitoring and technology with world-class data science experts to develop advanced technology solutions for transaction monitoring systems. The experience of working with clients around the world has helped KPMG’s technology solutions evolve rapidly to accommodate regulatory requirements while addressing client challenges.

• Risk-based - Customer segmentation can be more accurate and guided by the customer’s risk-based approach.
• Effectiveness and efficiency - AI tests thousands of different "what-if" scenarios and makes recommendations based on them for optimal calibration of customer segmentation, rules, scenarios, and thresholds. The optimised system greatly reduces the rate of "non-suspicious" alerts.
• Cost reduction - Optimised transaction monitoring systems generate fewer "false" alerts that require time-consuming manual review. Automated Level 1 verification minimises the requirement for the 4-eyes principle, depending on the risk appetite.
• Regulatory compliance - Automated detection of "suspicious" alerts enables faster referral of high risk cases to Compliance. Documented, transparent and auditable system testing and alerting processes demonstrate compliance with regulatory requirements.
• Insightful - Management information and analyses identify linkages, patterns and behaviours to provide real insights.