The Best Way to Automate Sanctions Screening

In calling out the need for automated sanctions screening, PwC recently noted that financial institutions (FIs) continue to face the potential for billions in fines for violations of US economic sanctions. PwC also observed that, despite increasing regulatory scrutiny and sanctions risk exposure, many FIs are still using outdated technologies and inefficient manual processes that result in inaccurate sanctions screening results.

Sanctions Screening Alert Fatigue

The problem starts with the rules-based sanctions screening software (which pretty much every FI uses). It generates a large number of sanctions alerts, and unfortunately, 98% of those alerts are false-positives. Still, each alert must be reviewed in order to discover the small percentage that are true-positives and pose a risk to the organization. That means manually reviewing each alert.

In the case of payment transaction alerts, a person must review each alert for identifying people, corporations, locations, and vessels from payment messages and compare them against the alert and external sources to recommend whether the issue can be resolved or escalated. Each alert review consumes 3–5 minutes of a reviewer’s time. With the multiple types of sanctions-related alerts and millions of transactions occurring each year, it comes as no surprise that FIs have built up large teams of people to manage the manual resolution of sanctions and transaction alerts.

In today’s economy, where new employees are expensive and difficult to find, it’s unrealistic for banks to continue hiring massive numbers of sanctions screening analysts in support of compliance around AML/BSA/OFAC/etc. Besides, doing so only adds to the high cost of manual processes which should be automated in the first place.

Intelligent Automation to the rescue

Using Intelligent Automation, banks are streamlining their sanctions screening processes. They are progressing from a four-eye maker-checker model to one where automated Digital Workers are the makers, and human experts are checkers who focus on new insights and quality control.

Working side-by-side with their real-world colleagues, WorkFusion’s pre-trained Digital Workers with machine learning (ML) review all alerts for false-positive scenarios and conduct research to justify their decision-making, validating each scenario. When finished, they then document the rationale by writing it in natural language to provide an audit trail.

Meet Evelyn and Tara, pre-built sanctions screening analysts

Evelyn Entity Sanctions Screening Analyst
Evelyn, Sanctions and Adverse Media Screening Analyst

One of the more popular Workfusion Digital Workers, Evelyn performs sanctions screening, including name screening alert reviews. This involves reviewing alerts for screened organizations and associated individuals. Here’s just a sample of the sanctions screening tasks a bank can expect Evelyn to perform — all on the first day of work:

  • Analyze and investigate sanctions alerts for entities, individuals, and securities
  • Disposition false-positive alerts based on multiple factors and escalate appropriate cases
  • Conduct searches, gather data and records evidence from internal systems, the internet and commercial databases
  • Accumulate facts from investigations for compliance and internal audit to ensure adherence to regulatory policies for customers and transactions
Tara, Transaction Screening Analyst

A similarly popular Digital Worker, Tara is a Payment Sanctions Screening Analyst that reviews sanctions alerts in real time to protect organizations from processing payments from sanctioned organizations and individuals. Tara also ensures fast and accurate processing of all transactions in order to provide a high standard of customer service.

How does she do it? Tara analyzes and investigates payment alerts generated against sanctions screening lists originating from free-format SWIFT and FUF messages. She scans internal systems, commercial databases and the public internet; gathers data; and records evidence, compiling facts for compliance and internal audits to ensure adherence to regulatory policies for customers and transactions.

As with Evelyn, Tara is far more than just an automated bot. Both Digital Workers are intelligent, trained and effective digital employees that work alongside your human colleagues to perform sanctions and transaction screening as efficiently and effectively as possible.

Sanctions-related screening will just keep getting easier with Digital Workers

WorkFusion Digital Workers are true knowledge workers that have been trained with data (securely) from multiple organizations and taught by subject matter experts in the tasks they automate. Their ML empowers continuous learning and improvement over time.

But they’re not just smart. They are fully equipped with a set of digital capabilities that will continue to improve the speed and accuracy of their screening processes. Beyond their ML that includes natural language processing (NLP), WorkFusion Digital Workers incorporate pre-built connectors to a multitude of systems, intelligent document processing (IDP), and many more features – all designed to work together to automate a multitude of processes.

At WorkFusion, we have taken the concepts of automation and digital workers to a higher level. Our AI-enabled Digital Workers are process automation personified. They are valued digital members of your team who work alongside their real-world colleagues to reduce manual work, enhance quality, increase speed, save money, support compliance, and expand the overall capacity of your team. They take care of the often mind-numbing and mundane — activities such as data collection, document handling, and false-positive clearing – freeing-up their colleagues to work on more strategic and fulfilling projects. And they expedite previously slow and ineffective work that helps to improve customer service.

To see how easily Evelyn and Tara can transform your sanctions screening processes in support of your AML strategy, contact us today for a demo.

Share this article