CUSTOMER RISK IN BANKING

ENHANCING RISK CALCULATION ACCURACY & SPEED IN RETAIL BANKING

Our team were challenged to support a regional banking group to become more reponsive in their apporach to computing customer risk

How we calculated risk 72% quicker with greater accuracy

Our customer needed a more robust and accurate assessment of its exposure to certain risks and regulations.

It feared that reputational harm, government fines, counterparty credit and liquidity risks could affect its financial health or solvency.

It therefore needed a series of risk analytics models that defined not just instruments, but positions and counterparties, along with market data and risk factors.

Almost as important was that it needed these insights quickly.

Leveraging Machine Learning to improve forecast accuracy

Challenge

The sector had undergone a series of regulation updates in the previous 15 months.

The Client was not satisfied in its’ ability to compute risk in a timely manner that would permit quick investigation and resolution.

The Client saw an opportunity to acquire and include a variety of data sources to support their efforts but had not arrived at ana analytical methodology to dleiver its’ objectives.

Solution

Our team supported the Client in integrating their data with updated third party credit information. We also worked closely with the risk team to design a simple process that would generate system alerts on individual cases.

A number of solutions were deployed to achieve the desired outcome, including:

  • Continuous event-based risk rating of entities, which performed continuous monitoring and sent automated, daily alerts on client rating changes
  • Risk attribute management to identify and rank high-risk clients for review, with flexible mapping and management of multiple data elements.
  • Case management and investigations included continuous monitoring of entire client relationship to provide a full analysis for reporting / audits.

Our client used 20 years of monthly account-level credit card data, credit bureau information and bank account information to better assess customer risk before granting loans or raising credit limits.

Accuracy increase in risk identification 

84% 

Reduction in risk calculation lead-time

72%