Overview
Our client, a California-based community bank, specializes in providing banking and credit services to SMBs in the region. They turned to us for help in mitigating risks as they looked to expand their commercial lending portfolio into new verticals.
Challenge
The client’s goal was to increase loan volumes in the food and beverage industry without adding to its risks. They approached us with a list consisting of hundreds of local restaurants. The client wanted to reach out to these businesses with preapproved credit offers but recognized the need to temper their outreach goals with more informed decision-making.
Lead prospecting in any industry comes with its own set of challenges. For banks, the credit risks associated with commercial lending compound these challenges exponentially. Misjudgments regarding business creditworthiness can easily lead to a portfolio filled with bad loans and consequential losses that exceed a bank’s set risk margins. In this context, conducting pre-screening due diligence on potential borrowers becomes crucial.
For our client, having an in-depth due diligence process for hundreds of businesses was a challenge due to the labor-and-time-intensive nature of the undertaking. Consequently, they made do with checking credit bureau and FICO scores, which by itself, do not provide a complete picture of business health. What they wanted instead was to establish an efficient pre-screening process that would produce faster yet safer, more accurate prospecting decisions, and help them limit losses to within acceptable margins.
TRaiCE Solution
Using proprietary LLM models, TRaiCE first collected all available data for each business on our client’s list. This was done by parsing over 71,000 media sources, available company reports, credit bureau reports, regulatory documents, etc looking for risk-related data points on each business. The in-depth investigation also included global and local news reports, social media, review platforms, online discussion forums, and consumer complaint/protection bureau sites. The platform gathered such information not only on the businesses but also its subsidiaries, owners, and guarantors.
Next, our NLP algorithms analyzed the gathered information to produce a time series risk scorecard for each business that signaled how healthy or unhealthy it was. To make decision-making easy for our client, we also ranked the list of restaurants they provided based on their risk status. All of this was achieved in 24 hours.
Findings
The TRaiCE platform assigns risk scores on a scale from -100 to +100, where a score of -100 indicates the highest level of borrower risk and a score of +100 represents the lowest level of risk. From our client’s list, TRaiCE found:
About 20% of the businesses had negative risk scores between -1 and -20 which put them in the medium to moderate risk category
The remaining businesses had positive risk scores of 25 or more putting them in the least risky category
Outcome
Using the TRaiCE platform, our client was able to establish an in-depth due diligence process that included all available risk variables. This was a step up from their previous routine and helped them better identify and remove risk-prone borrowers from their pool of potential customers. Consequently, our client had a lead prospecting list that was 20% - 30% cleaner. As an added bonus, the improved and auditable risk mitigation measures offered by TRaiCE made the final list more appealing to the bank’s final decision-makers. In this way, our client was able to expand into newer territories with more confidence and is now entrusting us with additional tasks of a similar nature.
