The Million Dollar Typo: Breaking Down Why Banks Spend on Ineffective Screening Programs


Recent reports shine a glaring light on the urgent need for new solutions that can help compliance teams more effectively identify and mitigate real risks at the same pace of doing business. Despite this ever-present need, banks spend an inordinate amount of time and money reviewing information unrelated to their customers when trying to assess customer risk. Support for that counterintuitive statement is embedded in the banks’ customer screening processes, which frequently deliver false positive alerts for manual review. Whether the root cause is unsophisticated legacy systems, an ineffective process or human error, many banks are still plagued by false positives and retread steps already taken to sift through this irrelevant information, operationally doubling down on wasting time and money.

With So Much Attention Being Placed on Technology and Innovation by Global Regulators, Why Does the “Status Quo” Persist?

Current daily screening processes are designed to flag potential matches to watchlist records, indicating customer risk for the bank to consider. Watchlist databases have long been the standard tool used by banks to identify risks related to sanctions and embargos, political exposure and certain adverse events. Customers are initially screened against the full watchlist upon onboarding and are then screened against incremental watchlist updates (e.g., a new event related to the watchlist subject, a new alias added to the record) each day thereafter. On the other side, if the customer information is updated (e.g., new address, updated name), that particular customer will be screened against the full watchlist again. Therefore, the incremental changes on both sides—the customer list and the watchlist—are captured on a daily basis.

The reasoning behind this daily screening process is sound, but the control execution is largely wasteful: Analysts spend the vast majority of their time reviewing false positives, including alerts they have already seen. Because the materiality of the incremental change is not commonly considered, banks often see re-alerts—alerts on a particular customer and watchlist record that were previously reviewed and discounted as a false positive. It is prudent to review a re-alert when new material information is available for either the customer or the watchlist record that may impact the decision on whether they match (e.g., a customer alias was added). However, the majority of re-alerts do not contain new material information.

For example, screening platforms commonly re-alert when a watchlist provider corrects a typo by adding a space or punctuation to a given record. The correction does not impact the decision previously taken to discount the original alert as a false positive, yet a re-alert is generated because the watchlist record was technically and appropriately updated. If that updated watchlist record was related to an individual with a common name or with many aliases, the result could create hundreds of re-alerts across the customer population for review. The number of re-alerts can be further compounded on the customer side if a customer is a legal entity with similarly named related legal entities or if the customer has relationships with the bank across several jurisdictions or business lines. In many banks, each of these re-alerts requires a two-person review (the standard “four-eye” check) by the banks’ screening alert reviewers, even though the same information was deemed a false positive by two previous people.

Punctuation corrections alone make the case for process enhancement consideration, but immaterial re-alert issues scale far beyond human error. There are a myriad of immaterial update examples to both the watchlist and customer record that do not impact the previous discount decision yet cause re-alerts for review. Immaterial re-alerts have been responsible for more than 20% of the total false positives at some institutions, leading to millions of dollars and countless hours spent globally reviewing the same irrelevant information over and over again.

Now Is the Time To Expect More From Your Screening Solutions

The re-alert problem is exacerbated by legacy screening applications that were originally developed with fewer data point considerations and unsophisticated cognitive computing capabilities. These legacy systems are unable to consider the breadth of customer and watchlist information available for match detection and therefore re-alert more often. For example, many of these systems are restricted to Latin script only. The inability to compare character-based languages (e.g., Chinese characters) severely detracts from the risk coverage objective but also causes more false positives due to the commonalities of the translated and transliterated names. In addition, legacy applications are restricted to name matching and cannot consider other important factors (e.g., date of birth, jurisdiction, gender, occupation, business nature). Instead, assessment of these factors is deferred to the two-person manual process, meaning unnecessary time and money is spent reviewing and repeatedly documenting false positive discount rationale such as “date of birth mismatch.”

Using all customer and watchlist data points available, technology advances are powering customer screening efficiency leaps. Next-generation systems can perform sensible name matching tasks including screening names in native languages and considering name origin in the matching logic in conjunction with broader automation including multi-factor discounting and record update materiality assessments. These advances can help reduce false positives, eliminate the re-alert problem and enable banks to redistribute their historically wasted time and cost to deeper risk analysis and mitigation.

With both regulatory complexity and risk scrutiny on the rise, advanced technology relieves the burden of unproductive legacy systems and unnecessary manual processes with timely and efficient risk detection. Next generation screening technology, not wedded to traditional methods, allows banks to stay within their risk appetite today while future proofing their program with technology that is by design eminently more adaptable. The adoption of new, scalable technology will contribute to a broader and deeper understanding of the ever-changing and unabating risk landscape, making banks much more effective in their fight against financial crime. Accelerating a bank’s ability to find real risk and make confident decisions about their customers and transactions is critical in their mandate as the gatekeepers of the global financial system.

Daniel Banes, managing director and APAC regional leader, Exiger

Vincent Wan, global head of data services, Exiger

The post The Million Dollar Typo: Breaking Down Why Banks Spend on Ineffective Screening Programs appeared first on ACAMS Today.