Sanctions screening used to be a quarterly chore: pull the list, run the names, file the report. In 2026 it is a continuous obligation. Lists are updated on no fixed schedule, ownership structures are designed to obscure control, and regulators expect you to keep pace anyway.
Why list matching is no longer enough
A plain name match against a single list will both miss real hits and drown you in false ones. Modern screening has to account for transliteration, aliases, partial matches and the fifty-percent rule -- the principle that an entity owned in aggregate by sanctioned parties is itself restricted, even if it appears on no list at all.
The dangerous counterparty is rarely on a list. It is owned by someone who is.
What good screening software does
- Consolidates sources. OFAC, EU, UK, UN and sectoral lists, refreshed automatically rather than on a human’s memory.
- Resolves ownership. Traces beneficial ownership so the fifty-percent rule is applied, not assumed away.
- Tunes fuzzy matching. Lets you adjust match sensitivity by risk tier instead of forcing one threshold on every counterparty.
- Rescreens continuously. A counterparty cleared in January can be designated in March; the tool should catch that without a human remembering to look.
- Logs every decision. Who reviewed a hit, what they concluded and why -- the audit trail is the product.
The false positive problem
A screening tool that flags everything is as useless as one that flags nothing, because analysts learn to dismiss alerts on reflex. When you evaluate a tool, ask for its false-positive rate on a realistic book of counterparties, and ask how alerts are tuned down without quietly hiding genuine risk.
Questions to ask a vendor
- How quickly are list updates reflected after publication?
- Is beneficial-ownership data included, or another subscription?
- Can match thresholds differ by counterparty risk tier?
- What does the audit log capture for a reviewed alert?
The bottom line
Screening software is not bought to generate alerts. It is bought to let a human make a defensible decision quickly and prove later that they did. Judge every tool against that standard, not against the length of its list catalog.