How to cut analyst time lost to KYB false positives

False positives are built into know-your-business (KYB) screening, which is why the real KPI is not “zero alerts” but how much analyst time each alert burns. In practice, the difference between a tolerable review queue and a compliance sinkhole comes down to workflow design, list quality, and how your screening engine handles repeat hits.

1. Industry research puts false positive rates in KYB and anti-money laundering (AML) screening between 90% and 95% of all alerts. McKinsey has reported that more than 90% of transaction-monitoring alerts at most banks are false positives, with only one or two in every hundred requiring action. PwC research places the range at 90–95%, while ACAMS estimates up to 90% for sanctions screening specifically. Duna’s own observations across customer screening queues suggest the figure sits closer to 95–99%, depending on vendor configuration.
2. The first reason is plain data weakness. Most KYB screening starts with little more than a name and a country, which makes it hard to distinguish one “John Smith” from another. The second is messy reference data: incomplete profiles, transliteration variants, and entries that have no practical relationship to financial crime. The third is fuzzy matching, which is deliberately permissive so it can catch spelling variations and partial data. Tighten it enough to remove false positives and you also remove genuine hits, which is a regulatory problem. So yes, the system over-alerts by design.
3. The operational answer is not to eliminate false positives, but to make each one cheaper. Allowing analysts to dismiss confirmed false positives in bulk from a list view, instead of opening a separate detail page for every alert, is usually the biggest time saver in a review queue. Showing the screened entity and the matched profile side by side at the list level cuts out the click-through cost on obvious dismissals.
4. Another useful control is to store a prior clearance decision and automatically suppress future hits on the same profile. The Wolfsberg Group recognises this under the labels “Good Guys lists” and whitelisting, provided it is documented and reviewable. For PSPs and merchants running long-lived relationships, that matters because the savings compound every time the same name comes back through screening.
5. Active sanctions designations can be encoded as policy-level knockouts, which removes ambiguity and leaves human review for genuinely uncertain cases. Artificial intelligence is part of the discussion too, but the constraints are the usual compliance ones: explainability, auditability, and zero tolerance for black-box decisions when a regulator asks why a hit was cleared.

The thing to keep in mind is that KYB screening is supposed to over-alert. The commercial question for PSPs, acquirers, and compliance teams is whether those alerts are structured so analysts can clear them in minutes rather than burning time on repetitive clicks and duplicate reviews.