Name Screening – Has the Industry Got the Process Right?

Published on June 12, 2018 by

Financial institutions are scrambling to get their ‘know your client’ (KYC) and anti-money laundering (AML) systems in place following changes to legislations and policies, in an effort to combat money laundering and terrorist financing.

Laws like the PATRIOT Act, the EU’s Money Laundering Directives and other global AML regulations require institutions to have robust and effective systems and controls in place so that terrorists, terror-financing organizations and other such individuals/groups can be identified and reported to the relevant authorities.

Due diligence (at onboarding and periodic reviews), previously associated with financial institutions, is now a key compliance prerequisite for other industries and sectors as well as they attempt to safeguard themselves from risk of doing business with unwholesome individuals and entities. Screening is an integral part of the due diligence process, and the scope of whom, what and when to screen is ever increasing.

Whom to screen: Traditionally, only a customer or vendor with whom an institution interacted directly was screened. Best practice has evolved to require screening of all related persons and entities:

• Owners – Direct, indirect and ultimate beneficiaries

• Authorized representatives (e.g., directors, signatories, and trustees)

• Counterparties (i.e., beneficiaries of a financial transaction)

What to screen: This has evolved from a reactive exercise to a preventive one. Initially, the objective was to identify (i) politically exposed or (ii) sanctioned or watch-listed individuals/entities. Current best practice includes flagging credible adverse news.

When to screen: Institutions are warming up to the reality that screening cannot be limited to a one-time or periodic activity; it needs to be integrated as a daily check or even as a real-time trigger.

Institutions have partnered with firms providing data and automated workflow systems to help them with the screening process, but it remains plagued with a high number of false positives (alerts where underlying transactions are not found to be suspicious), requiring deployment of manual labor to review and eliminate these alerts.

Challenges in implementing a robust and cost-effective solution:

• Data: Data is at the heart of any screening solution. Eliminating false positives requires accurate and complete data – not only information on the client/vendor, but also key data points (personal attributes, unique ID numbers and locations) of related persons and entities.

• Configuration: Screening solutions need to be configured optimally. If all required data attributes are input at the time of performing a search, the system should ideally not generate multiple alerts for review. Ideally, it should also auto close false positives. At present, a scoring process is deployed to identify criticality of an alert, and manual effort needs to be invested to distinguish false positives.

• Technology: Institutions need to be prepared to leverage next-generation technology.

What technology has to offer:

• Machine learning applications can read through multiple websites to identify, extract, and cleanse data and use it to eliminate false positives. Traditionally, users have manually accessed company websites and news articles to

  • Find key data points (personal attributes, unique ID numbers and locations) and

  • Prepare individual profiles (e.g., when and where a person worked or resided)

• Facial recognition compares profile pictures collected during customer due diligence (CDD)/enhanced due diligence (EDD) or other manual processes with pictures stored in screening databases.

Acuity Knowledge Partners aims to support clients in redefining the screening process by leveraging

• Resources effective in performing analytical research on unstructured data to gather otherwise ordinary events, news and associations/relationships

• Global privately and publicly owned data coupled with a real-time screening solution

• State-of-the-art next-generation technology frameworks (such as machine learning, artificial intelligence and natural language processing)

---

---