As AI makes it possible for anti-money laundering (AML) processes to become increasingly automated, efficient, and effective, rules-based transaction monitoring systems (TMS) are being supplemented with these solutions to drive down false positives, streamline investigations, and identify risk that is going undetected. This begs the question: what are the indications that your AML compliance program will benefit from advanced technology?
Inconsistent Decision-making and Reporting
The first indicator that advanced AML tech will benefit your compliance program involves program consistency. Using legacy processes forces individual investigators to independently choose what data is examined, how that data is interpreted for risk and how a case is reported. Combined, these independent decisions create inconsistent outcomes that make auditors and regulators uncomfortable.
Advanced technologies improve the three critical investigator metrics you need to monitor, quality, decision making, and timing. Checking quality determines if investigators are going through all necessary steps. Decision making is an assessment of everything done as part of the investigation, and whether a quality control check agrees with the decisions made. Timing naturally looks at how long it took the investigator to complete the investigative process.
Contrast independent investigator processes with the AI-enabled QuantaVerse Platform which automates data gathering, applies AI models for risk analysis, and then delivers findings in consistent, well-structured Financial Crime Investigation Reports (FCIRs). Analysts quickly examine risk scores for entities, counterparties and categories, and prepare narratives that can be used to explain why a case is cleared or a SAR filed. Without a solution like this, the risk of inconsistencies is high across all three metrics.
Your People are Finding Risk Your TMS is Missing
When determining if your AML program will benefit from advanced technology, consider where your SARs originate. If more interesting, better-quality cases are flagged by individuals throughout the institution noticing suspicious activity than those that are coming from your TMS, it indicates that your TMS needs assistance. It may not have the capability or configuration to capture things that are truly suspicious. For example, one smart bank teller can identify something that’s happening in all your branches that no one else is attuned to. While you can attempt to train everyone else to do what the smart teller does, it is more beneficial to train your TMS to do it, and you’re going to get better results.
Look for AI solutions that can automate detailed entity resolution and chain analysis of related counterparties. AI can also help tune the TMS and write new scenarios that you want your transaction monitoring to capture. What’s more, with efficient AI-enabled technology like the QuantaVerse Advanced Detection solution, all transactions are automatically considered, not just those above a certain value.
Prioritization of Alerts
Financial institutions put the alerts created by their TMS in some order of priority. Risks are typically ranked numerically where higher numbers correspond to higher risk. However, there are many alerts that score high based on arbitrary rules that do not necessarily indicate risk. Aggregating or “pinching alerts” that have triggered when an entity exceeds a threshold for wire activity, cash activity, and/or ACH activity will cause that pinched alert to have a score but not necessarily high risk. With the application of more data and better analytics, the true risk of these transactions can be evaluated.
Some institutions take their risk prioritization a step further by carving out higher risk activity, such as international wires. However, those investigations are extremely manual and often unproductive. Evaluating these cases often absorbs the time and attention of a subject matter expert to review the alerts initially before they are assigned to investigators.
The QuantaVerse Platform solves this prioritization problem. It automatically recommends which entities require no further investigations (around 75%), as well as those requiring further human investigation (around 20%) and the 5% that will likely need a SAR to be filed.
An advanced AML compliance technology, like the QuantaVerse Platform, also solves the assignment problem many investigative teams face. It helps managers instantly identify which alert should go to which investigator. For example, new analysts could be automatically assigned basic cases, with the more experienced investigators taking care of alerts that require jurisdictional expertise or in-depth investigation.
Your Alert to SAR Ratio is Under 1%
When your alert to SAR ratio is under 1%, it means that 99% of the time the alerts that you’re reviewing do not need a SAR. Although the AML industry has not solved this failure of transaction monitoring systems, it means you’re wasting up to 99% of your team’s time. An alert to SAR ratio between 5% and 10% is considered reasonable. But advanced AML compliance technology can deliver even better alert to SAR ratios.
Improving your alert to SAR ratio in effect means you’re wasting less time. So, look for an AI-enabled system that can identify scenarios that aren’t working, and eliminate the false positives that hurt your alert to SAR ratio.
Difficulties Resolving Investigation Backlogs When They Happen
Backlogs are a symptom of an underlying problem that normally has an identifiable solution. It could mean you don’t have enough people or that your TMS is not configured optimally. It could be the result of an unexpected event like the pandemic. If you’re never really resolving the backlogs, or you have a lot of difficulty resolving backlogs, it is an indicator that advanced technology will help.
An advanced system like the QuantaVerse Platform can easily reduce the number of cases that would otherwise contribute to your backlog. Having the right technology in place also helps you manage surges or ongoing growth with the team that you have by increasing investigative efficiency.
High-Risk Entity Reviews (HRERs) are Detracting or Absorbing the Valuable Time and Attention of Your Investigative Team
When it comes to high-risk customers, compliance teams get to know them well. They need to keep a deep file on each high-risk customer and be able to defend maintaining that customer to a regulator. All of this is a huge drain on the time and resources of your investigative team. In addition, most of the time you don’t want to overreact to alerted transactions involving your high-risk customers because you’ve already vetted them so well and you’re comfortable with what they’re doing.
A high-risk entity review takes up the lion’s share of the work that investigators must do on high-risk customers. Automating data gathering and risk analysis speeds up the high-risk customer review process by handing investigators risk scores and related narratives on high-risk customers. It saves the investigator from going through multiple systems to investigate if the high-risk customer is still active, if the news has changed, and if the year-over-year activity has changed.
Investigator Fatigue or Frustration
Human investigators spend approximately 70% of their time gathering information, and 30% conducting analysis. Investigator fatigue is a real phenomenon. So, any way you can whittle away at that 70% is worthwhile.
Automating data gathering, data organization and risk analysis combined with pre-filled FCIRs from the QuantaVerse Platform are a boon for investigator productivity. With most of the investigative work taken care of, human investigators are left to focus on more rewarding and self-motivating tasks.
As the industry goes through massive digital transformation and as regulators keep upping their definition of “reasonable” control and governance, next-generation AML is coming to the forefront. The barrier for entry has dropped to the point where it is within reach of even smaller institutions. You don’t need an army of data scientists on staff. Advanced AML technology is readily available to automate repetitive manual processes, more accurately detect suspicious activity, and cost-effectively put these capabilities in the hands of more financial services organizations.