96.5 F
San Fernando
Friday, Mar 29, 2024

Software, Data Equal Better Bank Loans

Fintech provider ZestFinance in Burbank has partnered with Digital Matrix Systems, a Dallas area risk management firm, to improve the way banks measure credit risk with loan applications. Bruce Upbin, a spokesman for ZestFinance, says it’s easy for credit data files to become corrupted, or fields are not filled out completely or have missing elements. “It’s good to work with partners who can clean that up on the front end so that the modeling goes faster, and we get a better result,” Upbin said. The two companies are combing Zest’s automated machine learning software with a Digital Matrix platform to make it easier to create and deploy credit attributes and scorecards for automated decisions. For Zest, the partnership means getting to market faster and an alliance with a name that a lot of banks already know. Additionally, it gives the banks confidence they will have secure, consistent and reliable access to clean data to build a Zest model, Upbin said. “It makes it easier for them to switch from traditional underwriting with a few dozen credit variables to machine learning underwriting which uses 10 times more data,” he added. Mark Dreux, head of strategy and business development at DMS, said the partnership will bring value to its clients, which include consumer and commercial credit bureaus, as well as specialty data providers. “The ability to pair Zest’s (machine learning) capabilities with DMS’ standardized and normalized attributes, as well as our clients’ custom attributes via our CreditAnalyst tool, will be a big win for lenders in support of their risk management efforts,” Dreux said in a statement. Working with a machine learning underwriting model means it requires a lot more data than banks are used to using. It can be tricky getting the data in the right format when getting it from the credit reporting bureaus, Upbin said. Digital Matrix is a specialist in aggregating and organizing the data the lenders need to make decisions on approving loans. Banks, credit bureaus and loan operating systems that smaller banks use to run their loan programs all use DMS products, he continued. “By partnering with people like DMS it makes it easier for banks to use our software,” Upbin added. ZestFinance was founded in 2009 by Douglas Merrill, former chief innovation officer at Google LLC, and a team of former Google employees. Its machine learning software creates a risk score, somewhat similar to a FICO score, but more nuanced and gives a more holistic view of the potential borrower, Upbin said. The software can help a bank improve its approval rates without added risk. Typically, the company sees a 15 percent increase. It also sees about a 30 percent reduction in credit losses that helps keep credit rates constant, he added. “That is the power of machine learning – better math and more data,” Upbin said. “Partnerships like the one with DMS help us get there faster.” “This partnership will remove a big challenge of doing (machine learning) — getting the right data in the right format,” Merrill, who serves as ZestFinance’s chief executive, said in a statement. “Together, we can help more lenders use (machine learning) to approve more good borrowers and control their underwriting risk.”

Mark Madler
Mark Madler
Mark R. Madler covers aviation & aerospace, manufacturing, technology, automotive & transportation, media & entertainment and the Antelope Valley. He joined the company in February 2006. Madler previously worked as a reporter for the Burbank Leader. Before that, he was a reporter for the City News Bureau of Chicago and several daily newspapers in the suburban Chicago area. He has a bachelor’s of science degree in journalism from the University of Illinois, Urbana-Champaign.

Featured Articles

Related Articles