Model Monitoring

Most lenders will expend considerable resource to create suites of internal models that support various Credit Risk functions ranging from acquisition and management of customers to the calculation of Regulatory Capital and Loss for reporting purposes. The ongoing assessment and maintenance of these models is integral to ensure they remain robust and fit for purpose throughout their life.

Traditional monitoring focuses on several metrics that are common regardless of the specific model:

With careful management and exploring the right metrics, the effective life of the models and the value to the lender can be maximised.

  • Does the model reflect the real-world observations, i.e. did the expected results from the model match what actually happened?

  • Does the model maintain the discriminative power of the original development? This is particularly important for scorecards and other multivariate decisioning models.

  • Are the assumptions and segmentation applied in the original development still appropriate and reflective of the current climate?

  • What degree of movement is expected and what actions should be taken in the event of breach?

The exact approach to answer these questions and the degree of automation and customisation varies by institution and model, from simple spreadsheets to processes that create custom summarised packs which highlight key areas, generate commentary and export and e-mail the results at the touch of a button.

We have been involved with over 20 IFRS 9 projects and have experience with every aspect of the design, development and validation lifecycle. Through this extensive experience we have considered and adapted our approach to fully represent the various moving parts of IFRS 9 in bespoke monitoring solutions.

These include:

  • Monitoring the granular month-on-month forecasted performance to better react to small shifts in the lenders portfolio and the industry, something that was impossible under the fixed outcome Capital models

  • Assessing the various overall ECL Economic driven forecasts over time. The majority of lenders use a weighted view of several Economic forecasts to create a balanced view. Monitoring and adjusting the weights of these forecasts as well as the forecasts themselves allows lenders far more control over their Loss reporting and provides management with a range of different forecasts using various macroeconomic scenarios

  • Understanding the live movements in collection and possession behaviour over time, allowing losses and forecasts to be adjusted with unbiased comparisons

  • Investigating the impact of potential Transfer Criteria adjustments upon the overall Stage flow rates and reporting numbers.

Our extensive IFRS 9 experience and interactive approach to working directly with Risk, Reporting and Management functions allows us to provide the most customisable and appropriate Monitoring solutions that are constantly evolving to keep pace with a dynamic industry.


For further information on Model Monitoring, please contact our Partner, Rob McDowell at


Latest Insights