Reduce false positives in AML screening without clearing the same match every week
Clear each match once, and keep sanctions screening tight enough to catch the real hits.
Trusted by 130+ financial institutions across 30 jurisdictions
A false positive alert in sanctions screening happens because a customer or payment detail resembles a sanctions, politically exposed person (PEP) or adverse-media entry closely enough to match. You can't reduce false positives in AML screening by turning the volume down, because loosening the matching means missing real hits.
Stop clearing the same match twice with Salv. Good-listing remembers what your team has already cleared, and clearance rules close alerts automatically on conditions you set, with every cleared alert kept on the record. Salv screens and monitors around 96 million transactions a month.
Why the same sanctions screening false positives come back
Most screening noise isn't new. It's the same customer matching the same list entry every time your book is re-checked: a common surname collides with a sanctioned individual, your analyst confirms it isn't them, and the next day someone confirms it again. Until each alert is resolved, the payment behind it doesn't move, so repeat clearing lands directly on your customers.
Clear it once: good-listing and clearance rules
Good-listing (what you may know as a whitelist) keys on identifiers you choose, such as a name plus an account number, so the same combination doesn't alert again. Clearance rules can key on person screening, counterpart name, transaction reference, or counterpart bank name or Bank Identifier Code (BIC). Take one from the library or write your own; one bank built a rule for a name sanctioned in one country, screened against customers in another. Test any rule against your historic data before it goes live, and re-screen customers only when their details or the list entry change materially.
Fewer false positives without missing real hits
The matching itself stays tight. Entries are normalised and transliterated, and you choose fuzzy, exact, contains or prefix matching per field, with a threshold you set. Lists come from Dow Jones, maintained by Salv and refreshed several times a day.
See it on your own data
The cleared alert is still on the record
Clearance rules close an alert after it's raised, so it's never deleted, and the rule that cleared it is shown against it. Months later, you can show a supervisor what was flagged, what cleared it and why.
Where AI sits
AI-Powered Insights adjudicates a name match and explains its reasoning, but it's informational only: it never closes an alert or decides.
What this will not do
Clearance rules only apply to automatic screening, not manual searches, and they can't make up for incomplete list data. Some entries carry only a name and alias, which limits how precisely a rule can be keyed.
AML tools that work together
-
ScreeningScreen customers and payments against sanctions, PEP and adverse-media lists, with matching you configure and lists Salv keeps current.
-
Risk scoringCustomer risk rating on weighted rules built around your framework. Kept current as the facts change, feeding the monitoring and screening that act on it.
-
Alert and case managementWhen an investigation grows, it becomes a case. Bring the related alerts, customers and payments together and work it from one place as a team.
Questions teams ask
An alert raised because a customer or payment detail closely resembles a sanctions, PEP or adverse-media entry without being it, usually through a shared name, transliteration or common payment reference.
Screening re-checks your book continuously against updated lists. Without a record of what your team has cleared, the same match alerts every time.
No. Good-listing only clears what your team has already cleared, clearance rules follow conditions you set, and the matching itself is unchanged.
Yes. It's closed rather than deleted, with the rule that cleared it shown against it.