A clean pass through the filter
The alert as it lands: Nothing reached your queue for this payment. A SEPA credit transfer from Lena Kowalczyk to Marta Vidal Interiors S.L. passed transaction screening automatically overnight: the filter generated one weak candidate, scored it well below threshold, and released the payment without human review. This case replays what the engine did so you can see why most payments never become alerts.
All people, companies, banks, list entries, sanctions programmes, and identifiers in this case are fictional and were invented for training. Any resemblance to real persons or entities is coincidental. The list data shown does not come from any real sanctions list.
Stage 1 of 5: Input
THE PAYMENT AS SCREENING SEES IT
- Rail
- SEPA Credit Transfer (SCT)
- Debtor
- Lena Kowalczyk (Poland)
- Creditor
- Marta Vidal Interiors S.L. (Spain)
- Amount
- EUR 1,840.00
- Purpose
- Invoice 2107 — interior design services
THE LIST (FICTIONAL)
- VIDAL MORENO, ErnestoAKA: Ernesto Vidal, E. Vidal MorenoDOB 1968-02-11 · Uruguay · Fictional Programme ALPHA
- MACHADO EQUIPMENT TRADING LLCAKA: Machado EquipmentUnited Arab Emirates · Fictional Programme KESTREL
Read the whole case as text (works without JavaScript)
Normalization
- Creditor name: “Marta Vidal Interiors S.L.” → “MARTA VIDAL INTERIORS” (Uppercase, strip punctuation, and remove the legal-form suffix (S.L.) so company designators do not distort name similarity.)
- Debtor name: “Lena Kowalczyk” → “LENA KOWALCZYK” (Uppercase; no diacritics, punctuation, or legal-form tokens to remove.)
Candidates
- le-vidal-alpha: matched on Shared surname token VIDAL — Token index lookup: the normalized creditor name contains VIDAL, which also appears in the list entry's primary name, so the entry is pulled in for scoring. The second list entry shares no token with either party and generates no candidate.
Scores
- le-vidal-alpha: total 0.26 vs threshold 0.85 (Name similarity 60%×40%; Entity type consistency 20%×10%; Country corroboration 10%×0%; Identifier corroboration 10%×0%)
Disposition
Cleared. The only candidate scored 0.26 against a 0.85 threshold: a single common surname token, an individual-versus-company mismatch, and no geographic or identifier corroboration. The engine released the payment automatically; no analyst ever saw it.
Sources for this case2
- Market practice
Wolfsberg Group Sanctions Screening Guidance ↗ — The Wolfsberg Group
Wolfsberg guidance is industry market practice, not law; institutions vary in how they apply it.
- Simplified educational illustration
Payments Signal editorial teaching models — Payments Signal
What this simplifies: Scoring model is a simplified teaching construct, not a production algorithm.
Used wherever diagrams, scenarios, figures, or example values are didactic constructions rather than sourced facts; every such use carries a simplifications disclosure. All people, companies, banks, and list entries in examples are fictional.