ZestFinance Debuts New Data Underwriting Model To Ensure Lower Consumer Loan Default Rates

Next Story

Exec Updates Its iPhone App For Ordering On-Demand Assistants, Adding Receipts, Multiple Jobs, New Sounds And Animations

ZestFinance, a company founded by former Google CIO and VP of engineering Douglas Merrill to legitimize the payday loan industry using machine learning and large-scale big data analysis, is debuting a new underwriting model today.

For background, ZestFinance takes an entirely different approach to underwriting by combining Google-style machine learning techniques and data analysis with traditional credit scoring. As a result, the company can help financial service providers better understand credit risk in their own businesses and better understand the creditworthiness of their borrowers. The ZestFinance decisioning infrastructure can run dozens of models in parallel, returning loan decision results within seconds.

Called Hilbert, the new underwriting model provides a more accurate depiction of a person’s ability to pay back loans, says ZestFinance. In fact, the new underwriting model offers a 54 percent lower default rate than the best-in-class industry score.

With Hilbert, approximately a quarter of the data that runs through ZestFinance’s underwriting models is based on new variables constructed by human predictive modelers. ZestFinance’s team of predictive modelers, who all have backgrounds in physics, computer science and mathematics, analyzes thousands of variables created by machine learning algorithms, modifies them based on patterns, trends, and unique insights, and feeds these variables into multiple big data models.

As Merrill explains to me, Hilbert increases the quality of underwriting analysis and is able to give more loans to people. The approval rate has doubled, he says. “This is a material step forward to give credit to the underbanked to save them money,” he says.

For example, in the case of bankruptcy (a strong signal used in underwriting decisions), a machine learning model can quickly calculate the number of years since a person filed bankruptcy, but wouldn’t know that people are only allowed to declare bankruptcy once every seven years. So, ZestFinance is making its algorithms smarter to be able to determine variables like this.

Since the company’s inception, ZestFinance has increased net repayment by 90 percent over industry scores and more than doubled the number of underbanked Americans the company serves. ZestCash previously raised $73 million in funding earlier this year.