Topic starter
23/06/2026 7:27 pm
TL;DR: Fraud programmes now hinge on correlating identity, device, and transaction signals rather than relying on isolated controls, according to SumSub. KuppingerCole’s Leadership Compass Report 2025 names Sumsub an Overall Leader in Fraud Reduction Intelligence Platforms, highlighting deepfake detection, behavioral intelligence, device intelligence, and fraud network detection for financial services facing account takeovers, payment fraud, and bot-driven abuse.
Editorial analysis by NHI Mgmt Group, based on content published by SumSub: “Sumsub's Fraud Prevention is recognized as an Overall Leader by KuppingerCole Analysts”.
Key questions
Q: What breaks when fraud controls only cover onboarding and not the rest of the user journey?
A: Stage-limited controls let fraudsters pass a clean entry point and then shift abuse into payments, payouts, or cash-out where governance is weaker.
Q: Why do identity, device, and transaction signals need to be correlated for fraud defence?
A: Each signal explains only part of the risk.
Q: How can security teams tell whether adaptive fraud detection is working?
A: Look for improvement in both detection speed and decision quality under changing attack conditions.
Practitioner guidance
- Correlate identity, device, and transaction signals Build decision logic that combines onboarding confidence, device reputation, behavioural anomalies, and payment context before final trust is granted.
- Map fraud controls to the full customer journey Review which controls stop at verification and which continue through session and payment steps, then close the gaps where trust silently persists.
- Tune step-up triggers for behavioural drift Use behavioural deviation, device change, and transaction irregularity as step-up triggers when valid credentials are present but session trust changes.
Bottom line: Fraud defence now depends on linking separate trust signals into one decision model rather than relying on a single checkpoint.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
25/06/2026 3:28 am
Fraud reduction intelligence is a control architecture problem, not a point-product problem. The article’s core lesson is that modern fraud cannot be governed by a single trust decision at onboarding or payment. Identity proofing, behavioural scoring, device intelligence, and transaction analysis have to work as one risk model if teams want to see the fraud path, not just fragments of it. The practitioner implication is to design fraud governance around signal correlation, not tool ownership.
A few things that frame the scale:
- U.S. fraud losses are projected to reach $40 billion by 2027.
A question worth separating out:
Q: Should teams treat account takeover, synthetic fraud, and bot abuse as separate problems?
A: No. They are different fraud expressions but they often share the same operating model: one identity path, one device pattern, and one monetisation step. Treating them separately creates duplicated controls and blind spots. A joined-up fraud model is more effective because it sees the full sequence instead of isolated events.
👉 Read our full editorial: Fraud reduction intelligence platforms expose the limits of point controls