A recent discussion with a senior consultant from one of the top banks in the world revealed the following underwriting challenges:
- There are easily 400 inbound loan requests that come in a day. It is practically impossible with a team of 25 underwriters to validate these applications.
- We don’t have a good handle on underwriting – while overall loss rates are within our acceptable ranges, what we don’t understand is how to price the risk at the top of the funnel so we can grow our market.
That’s precisely where HALO helps. An underwriting solution like HALO is not a replacement to your existing underwriting process but complements it.
HALO creates a dynamic underwriting scorecard that constantly improves itself based on the continuous flow inputs and outputs and identifies other factors that lenders are not even considering. HALO allows lending institutions to stretch the boundaries of creditworthiness based on their appetite for credit risk.
We have heard from lenders that they are buying leads and making loans for applicants they shouldn’t, and are passing on leads for applicants they should. HALO’s machine learning algorithms build a scoring model by interrogating all the input and output attributes to perform Artificial Intelligence-based underwriting. HALO helps lending institutions get smarter and fund more of the right merchants and less of the wrong ones.
What data sources do HALO’s Artificial Intelligence algorithms use to make underwriting decisions?
A. The application source (marketing channel)
B. The number of times a lender has seen the same “lead”
C. The inputs self-provided by the applicant (and how that compares to the actual data lenders get on credit reports and bank statements, etc.)
D. Demographics (geography, industry, age, etc.)
E. The attributes lenders receive from the alternative credit score (from companies like MicroBilt)
F. The attributes lenders receive from sources like Experian for ID verification fraud
G. The cash flow and transactions from the applicant’s bank statements
H. The choice the applicant makes as to amount, term and payment
I. Type of bank account
J. Business Tax ID time of issue
K. Type of business entity (LLC, Corp, Sole Prop, etc.)
L. The actual payment performance of the client to whom a lender has given an advance.
“HALO ultimately helps lenders figuring out everything from “A-K” that serves as a predictor of loan performance for the “Lender.”
Zuci is revolutionizing the way software platforms are engineered with the help of patented AI and deep learning models. Learn more about Zuci at www.zucisystems.com
About the Author
Vasudevan Swaminathan is the President and Chief Consultant at Zuci. Vasu is a trusted advisor and business partner to clients, having the ability to grasp their vision for the software. Check him out at Vasudevan Swaminathan