Whilst President Donald Trump’s tweets have already been found to move markets, and having significant consequences for investors, his more recent pronouncements about economic sanctions against Turkey are likely to prove even more of a headache for compliance departments. Financial firms are obligated to comply with economic sanctions as part of their financial crime compliance – and in the case of Turkey, must not transact or do business with any of the entities that appear on the US Office of Foreign Assets Control Sanctions List.

This may sound straightforward, but the process of screening customers or ‘names’ against sanctions lists is anything but. Name screening is one of the critical Know Your Customer (KYC) and anti-money laundering (AML) activities. In addition to sanctions, customers need to be checked to see whether they are Politically Exposed Persons (PEPs) and / or whether there is any negative news that may indicate they are a higher risk customer.

In theory, it should be simple to match a customer’s name (and some other details) to a list which contains details of PEPs, sanctioned entities and those with associated adverse media so they can be flagged as high risk and dealt with accordingly. In reality, however, there are multiple problems and issues that can result in high rates of false alerts (both positive and negative), impacting not only efficiency but the ability to identify the bad actors.

Why is name screening difficult?
Like most things these days, this is a data problem…one which innovative technologies can help to solve.

  1. Customer data quality – Effectively matching internal customer data with external name screening lists and data sources relies in part on the accuracy, completeness, consistency and timeliness of internal data sources. If data quality is below par, matching accuracy will be lower – ultimately resulting in higher levels of false alerts.
  2. Reliance on official lists and third party data – As well as keeping internal ‘black lists’ of potential PEPs or other high risk entities, financial firms are reliant on accessing many sources of external screening data. Official data, such as sanctions lists, are not provided in a consistent format or with all the data attributes required to make an accurate match (such as birth dates). These lists are also updated at varying frequencies and suffer from some of the same data quality issues as internal customer data.
  3. Name or ‘entity’ matching is inherently difficult – Any sort of ‘entity’ matching – finding a common entity between two or more sets of data is beset with difficulties (usually related to data quality) but there are some unique problems presented by name data because of their variability and complexity. These include:
    • Phonetic similarity but different spellings
    • Transliteration spelling differences
    • Nicknames
    • Titles & honorifics
    • Truncated name components
    • Out-of-order name components
    • Initials
    • Names split inconsistently across fields

As well as these generalised problems, there are specific challenges associated with different types of screening. In the case of sanctions screening, there are multiple official lists (e.g. the OFAC sanctions list, the EU Consolidated List of Sanctions, HM Treasury sanctions list and the UN Sanctions list) which are released in different formats and updated at different frequencies. For PEPs, there are no ‘official’ lists so FIs usually rely on a combination of third party data providers as well as internal ‘black lists’ that are built up over time (and suffer from similar data quality issues as customer data). 

Adverse media screening can prove to be even more problematic because it involves enormous quantities of unstructured data involving multiple languages – the phrase ‘looking for a needle in a haystack’ springs to mind. Additionally, adverse reputational information may be difficult to find, requiring deeper searching tools than generally provided by popular search engines.

How can technology help?   
There are two key considerations in developing or selecting a name screening solution. 

Products featured in the RegTech Directory (rtdirectory.co)
External Data for Matching

Firms need to decide how to source the list data related to PEPs, sanctions and adverse media. Three main options exist:

  • going straight to the source data e.g. official sanctions lists
  • using a single third party data provider 
  • using a name screening platform that integrates a number of data sources and / or maintains a dynamic database of high risk names. 

We are seeing a significant growth in products which aggregate and enrich multiple external data sources.

Name Matching Technology

Traditionally, name screening has been performed using rules-based matching techniques which make exact matches on specific data fields or attributes and whilst this is the easiest method it has relatively low levels of accuracy. Accuracy can be considerably improved by using more probabilistic methods such as advanced fuzzy matching algorithms and even elastic search. Again, newer products are offering more sophisticated matching techniques to address false alerts and drive efficiency in the name screening process.

False alerts can further be reduced by enriching customer data to provide additional data elements to aid the matching process. Some RegTech vendors are using AI technology such as Natural Language Processing to extract information from unstructured data, such as a date of birth or occupation and then creating rules based on this information that further improve matching accuracy.

We believe there will be a further development in name screening products that look to integrate name screening more closely not only into onboarding processes but the ongoing monitoring and risk assessment of customers. Smart vendors will either develop robust and flexible platforms that cover multiple financial crime compliance use cases and / or will innovate to further reduce false alerts through more sophisticated matching technologies or data enrichment from additional sources.

We have conducted a market assessment of all the Name Screening products in the RegTech Directory which maps the vendor landscape and highlights the capabilities of the vendors in this market segment. Click here to download the report.

Comments are closed