How We Helped A Regional Bank Overhaul Its Commercial Lending Unit Through Human And Machine Learning Credit Models!
The bank was one of the leading regional players in the lending market. However, the commercial lending segment was performing poorly. They were losing market share quarter-over-quarter to comparable competitive banks.
ACHIEVE PROFITABILITY IN COMMERCIAL LENDING SEGMENT
Performed a complete overhaul of their approach to profitability, that currently left them breaking even—or in the negative. Created new paths to profitable commercial lending underwriting.
Simplified customer/client acquisition, applications and portfolios, and eliminated the time-consuming and redundant areas of underwriting assessments.
GENERATE NEW CREDIT SCORING ALGORITHM
The new credit scoring algorithm outperformed existing model, passed the stress-testing, and identified borrower eligibility—all with the goal of rapidly delivering results.
The regional bank was struggling to justify commercial lending segment due to declining market share. The bank was sustaining devastating loss in their commercial lending segment quarter-over-quarter.
Dextro Analytics’ custom-tailored, adaptive decision making and solution framing were ideal for minimizing their negative returns. Our team of decision and data engineers dug deep, to generate a lead-to-assessment and disbursement of loan models. Our team:
1. In order to accurately identify gaps, we assess loan products from the consumer perspective.
2. Pinpointed and detailed the significance of simplifying the product and messaging.
3. Integrated new data sources to build sophisticated credit scoring model, that remains well within regulations.
4. Built expansive algorithms that helped internal data science team to improve the model accuracy by 6%, which translated into millions of dollars in new loan opportunities.
1. Simplified the customer journey, especially the assessment process.
2. Established next-generation of real-time credit scoring and rating. Designed signature tool that captures insights from a wide range of innovative data sources.
3. Minimize negative returns by overhauling the risk assessment, and applying a strategic approach at all levels of loan management.
“At first, we were puzzled as to why Dextro was requesting non-model specific information to assess credit scoring model. We now realize why – they helped to see our segment much differently. They not only improved the credit scoring model but also the acquisition and consumer lifecycle journey. No wonder we are seeing immediate results.”
- Director of Risk Modeling
Our team would love to share the nuts and bolts of the study and our best practices.
Please provide us with your contact information.