Early warning system reveals financial distress 5 months earlier than normal

BSG worked with a South African bank on a system to enable prediction of financial distress up to five months earlier using machine learning, integrated with a data pipeline, built on a big data stack.

overview of the clients needs
  • Reduce unexpected losses across credit products in corporate and investment banking (CIB)
  • Automate manual process of identifying high-risk customers
  • Transition from reactive to proactive risk management
  • Bring in risk factors from other areas of the bank that may be indicative of distress
  • Save the bank money lost as a result of default
  • Develop a tool to identify financial distress earlier and more accurately than existing credit models
  • Leverage other internal and external data sources to improve distress prediction
  • Create a robust data pipeline to enable consistent, reliable outputs
  • Translate the tool’s outcomes into business terms stakeholders can clearly understand
  • Improve the identification of risk by four to five months
  • Proactive risk management to enable the bank to assist customers by restructuring their debt
  • Time and money savings through proactive suggestions prior to customers being distressed
  • Sustainable outputs as a result of a robust pipeline that transforms input data into business intelligence-ready data sets
Bsg Early Warning System Predict Financial Distress

In a first for the bank, BSG took data consolidated from various sources and fed it into a predictive analytics model, generating insights around default risk, which is then integrated into daily operations, enabling the bank to proactively manage financial distress

Do you want to save millions?

Through data modeling, we developed an early warning system that enabled our client to identify customers who were at risk of defaulting on their loans. Through this early warning system we saved our client millions of Rands. If you want to use the data you have to potentially save your company millions of Rands, get in touch with our data analytics team.

Related Case Studies

Contact Us

We love talking about change. If you've been inspired by anything you've read here, please get in touch.

Not readable? Change text. captcha txt