How to Reduce AML False Positives


Meeting anti-money laundering (AML) compliance standards is a challenging operation for many organizations. With rules and regulations constantly increasing, it is vital for compliance professionals to streamline processes to free up time to focus on real risk.


One of the most inefficient parts of a compliance professional’s day lies in having to review false positives. 


This article will focus on ways your compliance team can reduce AML false positives to ensure that effort and funds are directed to an efficient and effective compliance program


In this article, we will take a look at the time-saving opportunities and effectiveness of advanced analytics, artificial intelligence (AI), rules-based analytic programs and other tools and strategies for reducing false positives. 




False Positives in AML Compliance

A false positive in AML compliance generally refers to when a transaction or customer record is flagged as potentially matching a name on a sanctions, watchlist or politically exposed persons (PEPs) list, but upon further review, the match is found to be incorrect. 


False positives are distinct from true positives, genuine matches to sanctioned entities that must be prohibited from transacting, and true negatives, which correctly clear non-sanctioned entities. False negatives, where a sanctioned entity is incorrectly cleared, are the most severe screening failure.


Businesses must strike the optimal balance between detecting true positives and minimizing disruptive false positives and false negatives.




What Causes False Positives?

There are several common reasons why sanctions screening processes generate false positives.




Similarity in Record Fields

Many sanctions lists contain minimal identifying information beyond the name. Thousands of sanctioned entities may share the same name as legitimate customers, leading to false hits based on name similarity alone.


Coincidental matches between sanctioned names and legitimate customers are especially likely for common names and names from regions with naming conventions less familiar to Western screeners.




Incomplete Data

When customer data or sanctions list records are missing critical fields like date of birth, address, or national ID numbers, the screening system has a harder time confirming a match. Missing data means fewer data points to differentiate a false match from a true one.




Outdated Data

Outdated or inaccurate data in customer databases or sanctions lists also causes false positives. For example, if a customer has changed their legal name but the system still uses their old name, it could trigger a false match.




Inadequate Contextual Information

Basic sanctions screening may not consider secondary identifiers and context that could distinguish a legitimate customer from a sanctioned entity. Details like associated companies, employment history, business networks, and geolocation provide valuable context to clear false matches but are often not factored into screening.




Overly Rigid or Inadequately Tuned Matching Algorithms

Sanctions screening systems rely on matching algorithms to detect potential hits. However, some algorithms are excessively rigid, generating matches on broad name similarity without considering other contextual factors.


Screening systems must also be carefully tuned and tested to fit an institution’s customer base and risk profile. Poorly calibrated systems generate excess noise.




Can False Positives Be Eliminated?

Sanctions screening false positives can never be entirely eliminated, primarily due to factors outside screening providers’ control. 


  • Sanctions lists are inherently limited in the identifier information they provide.
  • Sanctioned entities actively try to disguise their identities.
  • The matching algorithm must be tuned for high sensitivity to avoid false negatives.


They can, however, be reduced and limited.




The Inherent Difficulty in Reducing AML False Positive Rates 

Compliance reporting standards and regulations are currently set in a way that will always result in false positives to some degree. There are certain requirements, such as suspicious activity reports (SARs) or currency transaction reports (CTRs), that mandate reports be filed when certain thresholds are reached, such as cash transactions exceeding $10,000. In these instances, alerts are not subjective or open to interpretation, and therefore there are no ways to reduce the amount of CTRs you must file, even if they do not result in action taken by law enforcement.  In this case, gains of efficiency can be realized by simple automation, such as an automated regulatory reporting solution


Other areas of AML compliance, however, such as sanctions, watchlist and PEPs screening, if optimized with efficient rules, AI capabilities and scoring models can lead to higher operational efficiency and effectiveness of your compliance program.




How to Reduce Sanctions Screening False Positives

Although a perfect sanctions screening system is not possible, it is possible to substantially reduce the proportion of false positives. On average, businesses face a false positive rate of around 90%, which hugely inflates compliance costs. Even a moderate reduction in false positives and subsequent manual investigations can significantly reduce costs. 




Enhance Data Quality

Improving the accuracy, completeness, and timeliness of data in internal customer databases and sanctions lists is crucial. Higher-quality data provides more data points for the screening system to use to determine matches and reduces false positives triggered by missing and outdated information.


Sanctions lists should be updated promptly as they change. Customer data should be regularly reviewed and refreshed, especially for higher-risk customers. Collecting sufficient information when onboarding new customers is essential to enabling effective screening.




Contextual Data Analysis

Adding contextual data into the screening process helps filter out false positives. Screening systems can build a more holistic risk profile by using information like the customer’s nationality, occupation, account activity, locations, and transaction patterns.


With this context, the system can apply more sophisticated matching techniques to assess whether an alert is likely to be a true match or a false positive. For example, a transaction to a high-risk jurisdiction like Iran would be treated with more suspicion than one to a low-risk country.


Analyzing this additional context requires integrating the screening system with other internal and external data sources and applying AI/machine learning techniques to identify risk patterns. 



Sanctions, Watchlist and PEPs List Management

A startling reality is that many compliance teams lack the resources to properly review every match they get when conducting screening. We find that organizations sometimes have to turn off lists that they deem as “low risk” simply due to managing the high volume of false positives. The reality is that all sanctions, PEP and watchlists have some inherent risk, and turning off screening against certain lists can leave your organization susceptible to compliance failings.


We fully recognize that Sanctions and PEP monitoring can be extremely time-consuming, so we are seeing that many best practices stem from using data and technology together to save teams time and money while ensuring compliance.


Teams who have implemented holistic solutions are seeing significant savings in manual effort. This allows compliance professionals to regain time, and focus on matches that actually present a potential risk to their organization.




The Right Tools Can Help


Risk Scoring

Risk scoring assesses the likelihood of risk a potential match brings to your organization. Rather than a binary match/no match output, matches are assigned a probability score based on the strength of identifier matching and contextual risk factors. 


Risk scoring focuses limited compliance resources on the most impactful risks. A robust risk scoring solution paired with an AML case management system makes it easier for companies to investigate potential matches efficiently and easily resolve them.



PEP Scoring

PEP scoring models are another great tool to help reduce PEP-related false positives, going a step further than risk scoring, as they are designed specifically for screening matches. Scoring models allow you to still screen against all lists while freeing up time to allow teams to focus on their riskiest clients. These scoring models should be flexible enough to be based on your organization’s risk threshold and preferences, scoring models will automatically help assess and place PEPs into the appropriate risk category.


Most teams use categories such as low, medium and high risk, in an automated fashion, once again reducing the manual efforts required by compliance professionals. By using technology to categorize matches into differing risk levels, compliance professionals can tackle the highest-risk individuals first, and then choose to auto-resolve low-risk matches or review them at a later time. We need to be in the business of managing real risk not managing false positives, I know this is a bit of a preaching to the choir sentiment.




Advanced Matching Algorithms

Businesses should choose sanctions screening systems powered by sophisticated matching algorithms. Advanced rules-based matching improves match accuracy and can be tailored to the specific requirements of companies and industries. AI and machine learning models can be trained on an institution’s historical screening data to identify complex risk patterns and improve match accuracy. 




Rules-Based Analytics

As with most things, the key to being more efficient, and reducing false positives, lies in perfecting the basics. Advancements in machine learning and AI are exciting prospects for compliance professionals, however, these tools still need improvement, and unfortunately are useless if they are not implemented on top of an effective rules-based program.  


Identifying when a SAR must be filed or whether a client is a sanctioned individual, for example, requires effective rules-based alerts. 


For certain areas of AML compliance such as sanctions screening, the real danger is not missing out on reducing false positives, but rather implementing a program that results in false negatives, leaving your organization susceptible to fines and penalties from regulatory institutions. As a result, establishing effective rules-based analytics for your organization is a vital precursor before trying to reduce your false positives.


Running a rules-based AML program over time will allow you to analyze whether certain rules are needed. This can provide you with the data to customize your rules-based approach, removing unnecessary rules, to reduce the amount of false positives in your AML compliance program. 





Once you have effective rules-based analytics in place you can begin to improve efficiency with additional tools.


A combination of rules-based analytics with AI and advanced analytics may reduce false positives in AML compliance to a scale that saves your organization time and money.


The key here is to determine the potential benefits and ROI of implementing additional tools, such as AI. The real factor in deciding whether AI will add cost-saving benefits lies in an analysis of the scale of your operations and its compliance program, and whether the solution is accurate and intelligent enough to not cause false negatives.  


Large corporations will most likely benefit from additional tools, such as AI, due to the sheer number of transactions and clients they deal with daily. If fed with enough data, AI solutions can help to further categorize or resolve certain matches and alerts. A large organization utilizing only a rules-based approach will most likely yield too many alerts and matches for a compliance team to review. If implemented properly, AI solutions can help resolve large amounts of alerts generated through a rules-based program.  




Are AI and Other Advanced Analytics Tools Worth It?

To determine whether advanced analytics and AI will reduce compliance efforts and costs for your organization, it is first imperative to calculate your rate of false positives in your current rules-based AML compliance program.  


In some cases tweaking certain rules and removing others will allow you to reduce false positives without the need for additional tools or investment.




What Should Your False Positive Rate Be?

An ideal false positive rate will be dependent on the industry you operate in. It is estimated that most compliance programs have a 95-98% false positive rate. Having a robust rules-based system and AI capabilities could help to reduce this number. Even reductions of 5% could lead to a significant reduction in cost for organizations.  




The Cost of AI

Most organizations will find that the current costs of implementing additional AI tools will not provide higher savings than the initial costs of AI. For organizations in smaller and middle markets, the key is to establish an effective rules-based program, in addition to a cost-effective AI program, that meets compliance standards. Rather than spending funds to acquire costly advanced analytics and AI tools, focus on ensuring your current rules-based program is operating effectively with relevant rules.


As your organization grows, and clients and transactions increase, having already established an effective, customized, rules-based analytics program will provide you with the foundation to implement additional tools, such as AI, where needed. 


To learn more about rules-based and advanced analytics, view our in-depth look at the differences between these two compliance analytics 




Using Alessa for Improved Compliance Efficiency

At Alessa, we provide an all-in-one AML compliance software to help you meet compliance standards. Our solution includes real-time screening and monitoring of transactions, sanctions screening, automated regulatory reporting and more, and can be implemented as a holistic solution, or by specific modules per your organization’s needs. 


Our features, specifically our real-time transaction monitoring and screening and automated regulatory reporting, allow for improvements in efficiency for compliance teams and the reduction of false positives.




Watchlist, PEP and Sanctions Screening With PEP Scoring Capabilities

Our sanctions screening module is designed to reduce screening false positives by up to 50%. This is achieved through our proprietary PEP Scoring Model which categorizes matches into risk levels set by your organization’s risk appetite. 


Risk levels are based on:


  • Geography
  • Type of PEP
  • Riskiness of connections
  • Type of alert categories they fall into
  • Key data points


Learn more about our PEP Scoring Model here.




Real-Time Transaction Monitoring and Screening 

We mentioned above the importance of ensuring that your transaction monitoring system runs rules that are specific to your business. A second factor that can save your compliance team time and money is implementing monitoring and screening transactions that provide results in real-time. Allow your team to be proactive, rather than having to sift through old alerts that can accumulate over time. 


Alessa’s real-time screening and monitoring will also allow you to more accurately fine-tune your rules with the real-time data you receive. With Alessa, our users have experienced: 


  • 60% improvement in efficiency – leading to lower administrative costs
  • 70% reduction in alerts
  • 80% increase in alert accuracy




Automated Regulatory Reporting

Another great way to increase efficiency is to automate your compliance team’s reporting duties. Automating away time-consuming aspects of reporting such as creation, validation and e-filing allows your team to spend time in areas that require human decision-making and fine-tuning. On average, the implementation of our compliance software has allowed organizations to gain: 


  • 60% reduction in estimated time investigating suspicious transactions


This reduction of time and resources is accomplished by eliminating tedious manual data retrieval for standard requests and improving compliance teams’ workflows. 


Our rules-based analytics system allows for customization to ensure real-time monitoring of rules that matter to your specific industry and size. Paired with AI-driven sanctions screening capabilities, Alessa provides the foundational tools and data for compliance that grows as you do. Once installed, we update our software to account for changes in regulations and sanctions list so you don’t have to focus on minute changes made by compliance regulatory bodies. 


Contact us today, or get a free demo below to learn more about how our software solutions can help reduce AML false positives in your compliance program, allowing you to meet your compliance needs while cutting costs. 

Schedule a free demo

See how Alessa can help your organization

100% Commitment Free

Schedule a free demo

See how Alessa can help your organization

100% Commitment Free

Recent Posts

AML False Positive Rate Analysis

How to Reduce AML False Positives

Learn how to reduce AML false positives in your compliance programs and streamline your screening procedures to increase efficiency.

Please fill out the form to access the webinar: