How ComplyAdvantage Mesh drives faster customer onboarding by letting you screen against 49 types of risk

If you’ve ever built, bought, or used an anti-money laundering and countering the financing of terrorism (AML/CFT) software solution, then you’ll already be aware of the critical importance of high-quality data for successful customer screening and ongoing monitoring. 

AML systems rely on accurate, complete, relevant, and recent data to effectively detect suspicious activity. Fragmented or low-quality data poses serious risks for financial institutions (FIs), from floods of false positives that overwhelm systems and compliance teams, to case backlogs and missed red flags that allow criminal actors to infiltrate the financial system, resulting in severe regulatory and reputational damage to your business. 

ComplyAdvantage’s Customer Screening on Mesh represents the next generation of screening solutions, driven by proprietary data that leads the market in terms of depth and reliability. This article explains how fully integrated technology and data, along with advanced AI capabilities, enable our solution to deliver customer screening and ongoing monitoring with unprecedented levels of speed, accuracy, and efficiency. 

What makes data on Mesh different?

Unlike solutions that stitch together disparate systems or bolt AI onto existing architectures, ComplyAdvantage offers a single, horizontally and vertically integrated platform. In practical terms, this means: 

• Vertical integration: Unlike patchwork legacy systems, we own the entire data lifecycle from start to finish. We’ve built our architecture from the ground up specifically to manage complex financial crime data, ensuring seamless, end-to-end control from ingestion and enrichment to final resolution. 
• Horizontal integration: Our platform consolidates all your financial crime defenses into a single, unified system. By seamlessly linking customer screening, ongoing monitoring, transaction monitoring, and risk scoring, we create a single, holistic view of every customer and their risks. This powerful integration enriches data across all products, providing an unparalleled understanding of risk based on both financial crime attributes and behavioral patterns.

Using a single, unified platform gives you seamless, end-to-end control, while continuously enriching data to provide unparalleled insights and drive operational excellence. This architecture allows advanced AI to be embedded throughout the risk management lifecycle. Thanks to this architecture and these AI capabilities, ComplyAdvantage’s data independence is built on six pillars:

1. Data identification: ComplyAdvantage’s Financial Crime Compliance Strategy team leverages decades of collective experience in navigating the financial crime risk landscape and data complexities, bringing deep domain knowledge and practical insights to identify the data our customers need, jurisdiction by jurisdiction. This lets us provide granular risk categorization that meets our customers’ regulatory needs in the most efficient way possible. 
2. Data ingestion: ComplyAdvantage controls the ingestion process, using natural language processing (NLP) to read and categorize millions of unstructured data points in real time. This ensures real-time updates, as opposed to the delays and inconsistencies associated with external aggregators.
3. Data enrichment: Using advanced large language models (LLMs), ComplyAdvantage enriches raw data by extracting contextual insights and risk indicators. This goes beyond simple name-matching, providing a deeper understanding of potential threats by analyzing nuances in text and relationships.
4. Entity resolution: ComplyAdvantage’s sophisticated algorithms unify fragmented data, resolving multiple representations of the same entity into a single, unified representation. This minimizes duplicates and enhances accuracy to drive down false positives.
5. Data storage: Instead of siloed databases, ComplyAdvantage leverages a dynamic knowledge graph, continuously built and maintained by AI, to enable interconnected views of entities and relationships. This provides a holistic, 360-degree understanding of risk.
6. Search and monitoring: ComplyAdvantage’s AI-driven search capabilities facilitate real-time screening and automated monitoring, providing actionable intelligence and rapid alert generation with probabilistic scoring. A continuous feedback loop, driven by machine learning, refines detection algorithms and ensures ongoing accuracy improvements.

The new risk categories you can screen against with Customer Screening on Mesh

Thanks to this end-to-end proprietary data ownership and financial crime expertise, Customer Screening on Mesh gives you the ability to screen against seven top-level types of risk: 

• Sanctions exposure: This includes directly sanctioned individuals or entities, as well as entities associated with them by means of ownership, control, family connections, intermediaries, or enablement of sanctions evasion. 
• Political exposure: This refers to individuals who hold (or have held) prominent public functions, plus their relatives and close associates (RCAs). 
• Terrorist financing: This covers individuals and entities involved in the movement of funds to support terrorist organizations or activities, whether directly or indirectly. 
• Proliferation financing: This screens for customers involved in providing funds or financial services for the development, acquisition, or transfer of nuclear, chemical, or biological weapons and delivery systems. 
• Money laundering: This flags individuals or entities known or suspected of being involved in money laundering or any predicate crimes that generate illicit funds. 
• Regulatory and reputational risk: This includes those who have failed to comply with laws, regulations, or industry standards, such as those governing financial services or gambling. 
• Fraud: This encompasses individuals and entities involved in a wide array of fraud typologies, including identity theft, authorized push payment (APP) fraud, synthetic identity fraud, transaction fraud, and other related forms of fraud. 

When screening for any of these risk types, our system draws on relevant risk sources from: 

• Sanctions lists 
• Watchlists 
• Adverse media publications 
• Political exposure sources. 

However, within these categories, you can also screen against much more specific sub-types of risk. For example, suppose your business is particularly vulnerable to environmental crime, arms trafficking, or drug trafficking risks. In that case, you can opt to screen customers specifically for these risk types, rather than conducting a broader, more generic search that would inevitably yield more false positives. If you operate in an area known to be targeted by sanctioned entities for workarounds using shell companies, you can screen for indirect sanctions exposure via ownership or control.

In total, you can screen against 49 risk sub-types, allowing you to apply an unprecedented level of customization and granularity to your compliance policies and tailor them precisely to your risk-based approach. This precision-engineered approach to screening enhances both risk detection and the customer experience – allowing you to grow your customer acquisition rate without compromising on security. 

Our highly flexible risk framework also enables us to create new risk types for our customers to screen against as soon as we identify emerging trends or typologies in the criminal landscape, leveraging the benefit of our in-house risk and regulatory expertise. Whereas AML tools based on fixed risk schemas struggle to do this, especially quickly and at scale, our customers can use compliance solutions that move as fast as criminal actors do.

The State of Financial Crime 2025

Find out the latest financial crime trends and emerging typologies your firm needs to watch out for in our latest annual report, complete with expert insights and actionable tips.

Download your copy

How Customer Screening on Mesh uses AI to enhance efficiency in adverse media screening

In addition to these granular screening categories, Customer Screening on Mesh delivers significant efficiency benefits with advanced AI. 

Our adverse media screening solution uses LLMs to power risk detection, resulting in enhanced data ingestion and categorization. Because LLMs are highly capable of ‘reading’ media reports and publications, compliance analysts can obtain accurate, relevant summaries of adverse media content when an alert is generated, rather than having to manually review the content themselves. Using LLMs also means that the most important data points that determine whether an alert is a false positive or not, such as years of birth and categories of offense, can be far more easily and reliably extracted from an article. This allows for instant transparency into why a customer has been flagged, saving analysts time and letting them confidently make contextual decisions on whether to escalate alerts. 

Customer Screening on Mesh further improves screening efficiency with: 

• An enhanced matching algorithm, resulting in less noise from searches. 
• Fine-grained data source scope configuration, which allows you to further control noise in high-volume screening scenarios. 
• Fully integrated customer risk scoring with flexible rules. 

The result: an intuitive, AI-powered customer screening experience driving major reductions in both false positives and false negatives. 

The business advantage of integrated data and tech

With precision customer screening informed by financial crime expertise and based on integrated data, your risk management becomes significantly more flexible, allowing you to reduce noise and cut down false positive alerts. This has both compliance and business benefits, including: 

• Faster customer onboarding: With more straight-through processing now possible, you can create quicker onboarding experiences for customers with less friction. This helps you boost customer acquisition, reduce drop-off, and contribute directly to revenue growth. 
• Improved operational efficiency: Lower alert volumes and higher true positive rates mean your analysts can dedicate their time to the tasks that most need their attention, involving complex, high-risk decision-making. 
• Stronger regulatory outcomes: Regulators demand transparency, and unlike human-collated data or jigsaw-collected data, our AI-driven system provides a comprehensive audit trail from data sourcing to alert generation. Detailed data lineage and transparent scoring methodologies build regulatory trust and facilitate effective compliance. 
• Greater scalability: Our scalable architecture and deep AI integration ensure it can adapt to evolving threats. At the same time, our cloud infrastructure enables seamless scaling to accommodate growing transaction volumes and data demands.

“Customer Screening on Mesh represents a paradigm shift in how financial institutions are able to screen customers for risk. Legacy solutions have struggled to adapt quickly enough to changing risk landscapes and to deliver effective screening at scale without high volumes of alerts and false positives. Our granular risk categorization and LLM-first adverse media screening means we can.”

Paul Huston, VP Product, ComplyAdvantage

Scalable customer screening solutions for innovative firms

ComplyAdvantage helps fast-growing FIs achieve fast customer onboarding with real-time financial crime compliance, using precision intelligence to achieve both fast and secure customer screening. Customer Screening and Ongoing Monitoring on Mesh gives firms: 

• The industry’s broadest proprietary data coverage: Our solution covers sanctions, watchlists, enforcement actions, PEPs and RCAs, and adverse media, with over 500 million customers monitored annually. 
• Lightning-fast updates to risk data: Additions to global sanctions lists are available in minutes, rather than hours or days for competitors. 
• Fully flexible screening configurations: Users can set screening thresholds tailored to customer segments, locations, or business lines, and choose which risk sources to apply to different customer types or onboarding scenarios.