A New Approach to Building Credit Risk Models

January 18, 2017
By Patrick Reemts

The alternative data that we use in our credit risk model is constantly evolving to provide the best recommendations to our clients. Recently we’ve improved our approach to the way we build those models to better align our product development with the ever-changing needs of the industry by adopting a SCRUM Agile framework.

Agile certainly isn’t a new concept, it’s been around for years and many companies have adopted it for IT projects and software development, but we’re taking it beyond software development and applying SCRUM to all aspects of our product development.

We believe great products are built by a combination of market forces, technical innovation and scientific expertise, and in today’s environment, a dedication to transparency and legal best practices. At ID Analytics we now have multiple disciplines integrated and working on development together—from big data analytics, software engineering and data science, to our business and compliance teams. Everyone’s in the trenches collaborating together and sharing ideas in daily standups.

How does this benefit the companies that work with us? It allows us to not only evolve our credit risk model, but to efficiently tackle updates to all of the underlying data sources and technology that feed that model. And now, involve our legal and compliance teams early in the development process. It also lets us quickly iterate new modeling and methodologies for processing data. That means we are able to update our products more quickly to respond to the changing demands of the financial services and telecommunications industry. Our products will be more agile, if you will. We think this is a unique approach to building a credit risk model and one that we plan to apply to all ID Analytics products moving forward.

You can learn more about what makes our credit scoring model different here.

 

Patrick Reemts is Vice President of Credit Risk Solutions at ID Analytics, LLC