Join our Newsletter — 33% off our NHI Course

Are thin-file CRA checks turning good fintech applicants away?

 

(@nhi-mgmt-group)
Member Moderator
Joined: 2 years ago
Posts: 20230
Topic starter  

TL;DR: Thin-file KYC failures are often coverage failures, according to OneID, because CRA-only checks cannot find enough independent records. Onboarding needs broader evidence sources to avoid false negatives and recover genuine applicants.

Editorial analysis by NHI Mgmt Group, based on content published by OneID: “Crypto KYC: recover customers who fail CRA checks”.

By the numbers:

  • The FCA's Financial Lives survey put around 0.9 million UK adults as unbanked, down from 1.3 million in 2017.
  • The FCA found 22% of the 15.3 million people who applied for a regulated credit agreement in the two years to May 2024 were declined.
  • In one Tier-1 gaming operator's data, close to 30% of new customers failed a CRA-only 2+2 check.

Key questions

Q: How should fintechs reduce false negatives in digital onboarding?

A: Start by measuring how many real applicants fail the first pass, then add independent sources that can recover them without lowering assurance.

Q: When should teams move from a single-source check to a second wash?

A: Use a second wash when the first check fails a meaningful share of genuine applicants and the failure pattern is concentrated in thin-file segments.

Q: Why do good customers fail fintech KYC checks?

A: Good customers fail when the onboarding model depends too heavily on one identity dataset, usually a credit bureau file.

Practitioner guidance

  • Broaden onboarding evidence sources Add independent data sources beyond CRA files, including banking, mobile, insurance, and public-sector records, for applicants who fail the first pass.
  • Segment thin-file applicants explicitly Separate low-footprint customers from suspicious identities in your decisioning logic so manual review and alternative verification are only used where they are actually needed.
  • Measure false-negative onboarding loss Track how many legitimate applicants fail the first identity check, how many return, and how many open accounts elsewhere after abandonment.

Bottom line: Thin-file KYC failures are often coverage failures, not proof that the applicant is suspicious or unverifiable.

Glossary terms: Thin File Customer, Second Wash, Reliable Independent Source, Customer Due Diligence

Explore further

View Full Forum →  |  NHI Foundation Course →  |  Our Services →  |  Read the full analysis →


This topic was modified 2 weeks ago by NHI Mgmt Group

   
Quote
(@mr-nhi)
Member Moderator
Joined: 6 months ago
Posts: 20712
 

Identity coverage, not identity doubt, is the hidden failure mode in thin-file KYC: the problem is that many regulated onboarding stacks still equate absence from one credit dataset with absence from the market. That assumption was always narrow, but it becomes commercially damaging when the target customer base includes younger adults and recent movers. The implication is that identity governance for fintech onboarding has to be judged by evidentiary breadth, not by bureau match rate alone.

A question worth separating out:

Q: What is the difference between customer identification and customer due diligence in AML compliance?

A: Customer identification establishes who the customer is, usually by confirming core identity details and supporting documents. Customer due diligence goes further by assessing risk, understanding the purpose of the relationship, and applying enhanced checks where needed. Identification answers whether the customer is real, while due diligence answers whether the relationship is acceptable and under what conditions.

👉 Read our full editorial: Thin-file KYC checks miss good fintech customers


This post was modified 2 weeks ago by NHI Mgmt Group

   
ReplyQuote
Share:

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.