Collection departments utilise diallers and collections management systems to improve their collections by segmenting the delinquent customers, prioritising them and applying a host of treatments. Whilst segmentation takes you so far, there are a host of other mathematical models that can be explored to improve what we call the “Collections Cascade”. Improvement in any step of the cascade can help improve the collections yield and there are a number of models that can be used. A few of them are listed below:
Where to start
If your collections department doesn’t use scores, a reasonable question might be, “where do I start?” Typically, this would be with pure collections scores such as “Probability of missing next payment” (also known as roll-cards) and “payment projection scorecard” for late-stage collections (cycle delinquent ≥ 3months). Pre-delinquency is also a hot topic with some collection departments opting for a straightforward SMS while others offering a monthly lucky draw for a car if you make your monthly payment! Most recently in South Africa, connectivity has been a regular problem that collection managers have raised. That is where the probability of contact and right-time-to-call models are useful.
At Principa, when we look at any analytical solution we have adopted a “3-D approach” – determine, develop, deploy. Determine the business problem and data; develop the model; then deploy it. We consider all three when we start a project. A model is useless unless it can be deployed. For collection scoring, Principa utilise DecisionSmart to deploy our collections models.
DecisionSmart is a business rules management system that allows the user to deploy complex scorecards and collections strategies.
These can be fed directly into a collections management system or dialler. Principa have also integrated DecisionSmart with CollectSmart so that scores are calculated directly from the loans processing system and sent directly into CollectSmart for smart segmentation. If you have adopted machine learning, we can even use DecisionSmart to call a Python or R machine learning model through a SQL stored procedure. You can find out more about DecisionSmart on these useful pages:
To find out more or to organise a demo of DecisionSmart, please contact us here.