Credit Scoring

We use proprietary credit-scoring models that are trained on large, diverse datasets. These models learn to assess the credit risk of a counterparty - for example, estimating the probability of default or assigning a rating class - by identifying patterns and relationships within the data they were trained on. When you submit a scoring request, the model applies this learned knowledge to your specific inputs and generates a credit-risk estimation for the counterparty at hand.

Our models are trained on high-quality data, which is essential for producing performant and robust results. Data quality for us goes beyond the raw inputs and includes the entire data processing pipeline. After training, we conduct extensive out-of-sample tests - including evaluations across different datasets - to verify the predictive accuracy of each model. No model is perfect, but the validation scores we achieve provide strong evidence of their reliability. We also benchmark these results against what is realistically achievable for the given scenario. Our top-performing models reach AUC values above 90%.

Yes. We can assess counterparties of various profiles, and we may apply different scoring models depending on the specific context. To ensure the best fit, we typically recommend an introductory call so we can better understand your needs and the environment you operate in. If we determine that our scoring solution is not suitable for your particular use case, we will discuss this with you openly and directly.

In principle yes, but it depends on the quality and quantity of the data. If the data is not sufficient, we would rely on an existing data source instead. If it is sufficient, we can build a custom model and benchmark it against other models to ensure you can make the best possible decision.

Our pricing is determined on a case-by-case basis. We aim to offer the most efficient pricing for each client by considering the specific use case, the scope of the client's operations, and the expected frequency of usage. This approach enables us to provide competitive and fair pricing.