KPMG firms are plotting a new course for sanctions compliance with deep industry understanding and solutions that drive efficiency and effectiveness.
The pressures of financial crime are not new, but they are growing. The current challenge for banks is a combination of ever increasing and evolving regulations and legislation. These create a complex regulatory environment to operate in and comply with. Getting compliance wrong is not an option.
Existing sanctions screening systems often result in the production of large volumes of false positive alerts, often in excess of 95%. Traditional tuning techniques, applied to sanctions screening systems, require constant refresh due the increasing number of payments, customers, and the introduction of new sanctions regimes and frequent changes to existing lists. The limitations of poor systems tuning and regulatory change have resulted in alert spikes and backlogs in many institutions.
Current alert review processes frequently rely on extensive teams of personnel, often spread over disparate geographical locations. Delays in processing transactions, due to human processing times, can impact on customer experience, which is increasingly at the forefront of Financial Institutions strategic agendas.
In the face of increasing and evolving regulation, increasing transactional volumes, and the introduction of new sanctions regimes, KPMG has developed solutions that target both systems optimization and alert review processes. The goal is an increase in quality of screening systems performance and faster, more accurate and consistent alert review, helping Financial Institutions to meet their sanctions obligations more efficiently.
Sanctions screening testing platform
KPMG's Sanctions screening testing platform enables clients to test the effectiveness of their customer and/or payments screening systems.
- The Sanctions screening testing platform has the capability to use lists from Regulators, third party list providers or client's internal lists. The testing approach can follow the standard KPMG methodology or be tailored to a client's perceived risk.
- Results are analyzed with clients to establish the root cause of non-alerting tests in addition to identifying opportunities to optimize their screening solution.
- Files are generated in minutes rather than days, with an auditable trail of rules incorporated and lists used to determine test file composition accuracy.
The sanctions screening testing platform provides a robust and auditable way of testing screening system effectiveness and efficiency whilst reducing the potential for human error and cost of systems review screening testing platform
KPMG Sanctions Alert Classifier
The KPMG Sanctions Alert Classifier automates alert review decision making at the first level of review. It can deliver 99.998% accuracy in alert review decision-making with a variance of 0.001%. Alerts can be decisioned faster, with a consistently higher degree of accuracy at reduced cost.
- Using advanced machine learning techniques, the tool automates the identification of 'True' and 'False' positive Level 1 alerts, at a rate of up to 1m alerts per minute. The tool can be used with all sanctions screening systems currently in the market.
- Each alert review decision has a confidence level and is supported by a human readable decision rationale. Clients can therefore tailor deployment of the tool to dictate coverage and accuracy rates that support their risk appetite.
- The use of supervised machine learning ensures the tool is transparent and can be independently reviewed by auditors and regulators
KPMG has combined expertise in Financial Services, Sanctions and technology with leading data science professionals to create advanced technology solutions for sanctions screening systems. Key benefits include:
- System effectiveness. Fewer alerts produced, reducing impact on alert review teams
- Increased and consistent accuracy. Results of 99.9% accuracy of alert review based on millions of SWIFT messages vs. average human review accuracy of 92%
- Regulatory compliance. Automated 'True' positive detection enables faster escalation of high risk cases to compliance. Processing power of up to 1 million alerts per minute prevents build up of alert backlogs and effectively addresses alert spikes
- Cost reduction. Automated L1 alert review of up to 96% of alerts generated minimises the requirement for 4-eye human review
Improved customer experience. Timeliness of alert review reduces the delay in processing transactions.
Let's do this.