Screening a customer against a sanctions or politically-exposed-persons list sounds like a lookup. In practice it is a fuzzy-matching problem: names transliterate differently across languages, dates of birth are sometimes missing, and a screening engine tuned too loosely buries analysts in false positives while one tuned too tight misses genuine matches.
We build screening systems for onboarding and ongoing monitoring against OFAC, NACTA and equivalent sanctions and PEP data sources, tuned for your actual customer base rather than a generic default.
What We Offer
Name matching and fuzzy logic
Matching algorithms tuned for the transliteration and formatting variance in your actual customer data, not a naive exact-match or an untuned fuzzy match that floods analysts.
Onboarding and ongoing screening
Screening at customer onboarding and on a recurring basis against updated lists, since a customer who was clean at onboarding is not guaranteed to stay that way.
List management
Ingesting and updating sanctions and PEP list data from your chosen data provider, with the screening engine checked against known test cases after every update.
False positive tuning
Continuous tuning based on real dispositions, so the screening engine gets more accurate as your analysts confirm or reject matches, not static from launch.
How We Help
The operational cost most institutions underestimate is analyst time spent clearing false positives. A screening engine that is technically working but poorly tuned can generate ten false alerts for every genuine one, which either burns out the compliance team or, worse, trains them to clear alerts without reading them properly.
We treat tuning as ongoing work, not a one-time calibration. As your customer base and the sanctions lists themselves change, the matching thresholds that worked at launch drift, and we build the tooling to monitor and adjust that over time.
Our Approach
We tune matching thresholds against your actual customer data’s naming conventions, not a generic benchmark, since transliteration patterns vary significantly by region and language.
Every list update runs against a set of known test cases before going live, so a data provider’s format change does not silently break matching.
Technologies We Use
Industries We Support
Related case studies
- goAML-Integrated AML Monitoring for a Tier-1 Bank: Screening, monitoring and goAML reporting rebuilt around an immutable transaction event stream, so any alert can be reconstructed exactly as the system saw it.