Debt Collection

How Business Intelligence AI technology can help optimize debt collection.

The Challenge

The Debt Collection department of the client had very limited resources and their effort was not always worth it. They were not able to foresee the results of their debt collection effort or the time spent on a single account.
However, they had a database with a long history of debt collection experiences together with the details of the actions taken and the results of the actions.

Sigma’s classifier learns from the historic and current data, so it can later estimate the probability of debt recovery for each debtor.

The Solution

Sigma helped organize the historic and current data, so that Sigma’s classifier learnt from the examples and clustered debtors in categories, which allowed the implementation and optimization of a cost- and time-effective debt collection strategy.
The learning process takes less than 2 hours once the data is available.

The Outcome

The accuracy of the prediction of debt recovery probability was 80%. These predictions allowed to focus the limited resources on those debtors from whom the debt was more likely to be recovered.
They also had very stable monthly results and increased the profitability of the department.