## ## ##

 

CASE STUDY

A Decision Engine for Customer Remediation

Automation technology to enable a UK retail bank to find and remediate past customers.

The Client

The client was one of the UK’s largest retail banking groups, comprised of a portfolio of well-recognised, high street brands.

The client - along with its competitors and other consumer finance businesses - had been selling a certain class of financial product widely and for a prolonged period. The Financial Conduct Authority (FCA), having identified a number of concerns around the selling of these products, asked every bank to review their entire book of business, remediating any customers who might have suffered financial detriment as a consequence of buying the product.

With dozens of affected products having been sold via many different sales processes, and with tens of millions of customers potentially affected, the scale and complexity of the required response to the FCA was intimidating.

THE CHALLENGE

The client had produced a lengthy suite of assessment policies for its various product classes and a detailed contract strategy process. They were confident that by following these policies and procedures they could demonstrate to the FCA that every past customer had been fairly treated and that all reasonable efforts to contact them had been made.

The client recognised that automation technologies would be required to complete a project of this scale and Unai were invited to develop the technology to manage the process. The Unai team was deployed as part of a mixed team, with client and other supplier staff working together on the programme.

What Unai Delivered

  • Data Validation Engine. Unai's data engineers developed a validation pipeline to integrate and quality-assess all data relevant to potential misselling from across the bank. Data that failed the extensive quality checks was rerouted for manual analysis by teams of analysts.

  • Remediation Rules Engine. The Unai team turned the bank's remediation policies into sets of human-readable rules that could be applied to the validated data. The team then developed a rules engine which could execute these rules on the data. The traceability from policy to rules to code proved that the policy was being implemented faithfully.

  • Workflow Engine. The Unai team then developed a workflow solution which could manage each identified case of potential detriment through the contact and remediation process. Every action taken for every case in the system was fully audited, enabling the bank to prove that all reasonable efforts to contact past customers had been made.

The Result

Unai’s Decision Engine ran for over two years in production and handled the tracing and contact process for over 3 million customers, to both the FCA and the client’s satisfaction. Estimates from the client suggested that, with manual processing and evaluation times requiring 4 hours per case, an operation of this scale would have required 3,000 staff over the same time period.

The remediation system was reviewed by both internal and external auditors and praised as a complete system of record for every decision and every activity that occured during this programme.