Risk-Based Screening Isn't a Compromise. It's the Standard

A rulebook that treats every customer the same isn't rigorous. It's just slower to fail.

Tamas Simon

CCAS, Compliance Officer

5 min read

Global AML, KYC, sanctions, and CDD penalties totalled roughly $3.8 billion across regulators in 2025, per Fenergo's tracking, down from $4.6 billion in 2024 and $6.6 billion in 2023. Almost none of those fines went to firms with no controls at all; most had a program. What they didn't have was a program that adjusted to the customer in front of it. A single, uniform check applied to every account is easy to defend on paper and easy to defeat in practice, because the people worth catching are the ones who know exactly what the uniform check looks for.

One screening pass isn't one screening level.

A jurisdiction can be clean while the person behind an account isn't, and a person can be clean while the address they're transacting with isn't. Running screening at all three levels, jurisdiction, person, and blockchain address, closes the gaps between them rather than assuming a pass at one level covers the others. That still leaves a real design question: how much weight should transaction velocity and pattern carry against static KYC data collected once, at onboarding, and rarely revisited unless something trips a wire? The honest answer most mature programs land on is that the two shouldn't be scored independently at all. Velocity and pattern are risk factors folded into the same weighted model as the original KYC inputs, re-run continuously, rather than treated as a separate, later check on a decision that was already made once.

Corporate structures hide as often as they reveal.

Retail identity verification is comparatively simple. A corporate or business account requires looking through the entity to its ultimate beneficial owners. That's the right instinct, but it doesn't resolve on its own: at what ownership percentage does a look-through stop being proportionate and start being an unworkable search for control no structure fully discloses? Most regimes converge on 25% ownership or control as the baseline threshold, tightening to as low as 10% for higher-risk customers. That's a useful anchor, but it only works if a firm can say plainly which of its own customer segments get the tighter number and why.

Every proportional response is a bet.

A risk-based model doesn't just flag; it's supposed to produce a response sized to the risk, continued monitoring at one end, a hold or escalation at the other. The standard answer isn't a single analyst's judgment call. It's a documented risk appetite statement translated into a scoring grid across customer, jurisdiction, product, and channel, with a "four-eyes" second review built into ordinary cases and a specialist committee reserved for genuinely contested or high-risk ones.

What this means.

Risk-based screening only earns the name if the risk model itself gets revisited as often as the customer base does, and if its thresholds are documented well enough to explain to a regulator without reconstructing the logic after the fact.

MORE FROM THE COMPLIANCE HUB

Stay current on how BTSE builds trust

Licensing milestones, partnerships, industry events, and the latest from BTSE's compliance and risk teams.

COMPLIANCE

HUB

Contact Us

Copyright © 2026 BTSE

All rights reserved.

COMPLIANCE

HUB

Contact Us

Copyright © 2026 BTSE

All rights reserved.

COMPLIANCE

HUB

Contact Us

Copyright © 2026 BTSE

All rights reserved.