Competition from new age players, have made flexibility and response time more important than ever. There is a strong need to quickly bring new and advanced capabilities to the commercial lending market to remain competitive.
Businesses are short on time and looking for easy, intuitive experiences to help them manage day-to-day finances. And they're increasingly willing to leave their traditional banks or credit unions to find those experiences. For example, Raddon survey, found 32 percent of businesses were extremely or very likely to use an alternative lender to meet future financing needs, up from 23 percent in 2014. And 51 percent of millennial-led businesses have applied for a loan from an online lender that is not a financial institution. This is a phenomenal shift in customer preferences.
Non-traditional lenders have gained good traction owing to their shift in leveraging technology to service markets that were not serviced previously and borrowers whose credit scores are being augmented with new & alternative data to arrive at a score that is more robust and a holistic reflection of their credit worthiness.
Commercial banks and credit unions must continue to position their organizations as open for business in the era where decisions are getting more and more customer centric. According to an Epsilon survey, 80% of consumers are likely to choose products which are more personalized. The headwinds are strong in this direction.
Banks continue to use regression models to underwrite loan applications. Bureau data combined with application information determine the credit worthiness of the client. Regression models being linear in nature, isn’t able to bring nonlinear features and relationships into the equation like the recent techniques like GBM, random forest, etc. have done.
The crux of the problem lay in the ability to ingest and make sense, efficiently, the large pools of unstructured data available out there that could aid credit decisioning. This calls for an automated process to marry structured data points with data points which are qualitative in nature to a larger decision point.
However, all this data and modeling innovations will reap benefits only if it is capable of being sensitive to the environmental changes. Data to decision cycles cannot be as long as it is now and it is imperative to deploy them at as much real time as possible.
Scienaptic addresses these problems with its platform Ether. It is an advanced AI powered underwriting platform with prebuilt features which can be deployed within 4-6 weeks and deliver drastically better decisions.