Customer risk rating

Decision: Banking — KYC & onboarding Pattern: Combined approach

Use case: This solution rates a customer's risk and sets their due diligence level automatically wherever it has enough information to do so — and where it doesn't, it asks for the missing information rather than declining or guessing, so most cases require no human in the loop at all. See the KYC / CDD Risk Rating & Perpetual KYC accelerator for a ready-built starting point.

End-user: KYC analyst — only involved on the cases Rainbird can't complete without further information

The decision: What risk rating and due diligence level applies to a customer, at onboarding and through perpetual KYC reviews.

The knowledge graph: Encodes the bank's risk rating methodology and due diligence policy.

Integration: Customer and case data feed into Rainbird automatically. Where everything needed for a rating is already present, Rainbird reaches its conclusion and the case proceeds straight through — no human involved. Where a specific fact is missing, such as a source of funds or a beneficial ownership detail, Rainbird asks for this additional information to be provided, rather than declining or making an assumption. That request can route to a KYC analyst's queue, or to an agentic system that can retrieve the answer directly from a registry or the customer. Once the fact is supplied, Rainbird resumes reasoning from where it left off and reaches its rating.

Evidence: Every rating — whether reached straight through or after a follow-up request — returns the exact facts and rules that produced it, so the audit trail looks the same regardless of whether a human, an agent, or no one at all was involved along the way.