When Digital Matrix Systems began offering risk management solutions to credit underwriters in 1982, the financial landscape looking a whole lot different than it does today. Interest rates that year peaked at 21%, inflation was over 6% and banks were failing at rates not seen since the Great Depression.
The landscape looks far better today but the role of DMS is just as essential as it was then: to help financial institutions use data to minimize credit underwriting risks. DMS helps its banking customers incorporate consumer credit data flows into their risk models by standardizing and normalizing credit bureau attributes. Essentially, DMS ensures an apples-to-apples presentation of consumer information from the five U.S and two Canadian credit bureaus.
Today there is exponentially more data than ever, making absorbing it, understanding it and capitalizing on it more pressing than ever. That’s a big reason why DMS is turning to machine learning (ML) and partnering with Zest AI. Banks need more powerful tools to help them make better underwriting decisions in such a data-rich environment.
“Data has become both more important and less important at the same time,” says Mark Dreux, head of strategy and business development at Digital Matrix Systems. “What I mean by that is one individual source of data has become less important, but combing through all of the data available and trying to use it – even just as a way to stay competitive – has become more important.”
"With Zest, we’re doing everything that fits with our mantra – connecting to data sources, making modeling faster and enhancing analytics.”
Last year, it became clear to DMS leadership that its customers needed a simple way to fold artificial intelligence (AI) and machine learning (ML) into their businesses to stay competitive. ML’s advanced math allows lenders to use hundreds of times more data points for every customer so they can make more accurate decisions about a borrower’s creditworthiness.
So DMS partnered with Zest AI to integrate Zest Automated Machine Learning – ZAML for short – into DMS’ Data Access Point, a platform that simplifies creation and deployment of credit attributes and scorecards for automated decisions, risk assessment, and probability calculations. DMS provides connections to more than 20 loan operating systems and has connections with more than 40 data providers and all major credit bureaus.
Before the agreement with DMS, Zest partnered directly with financial institutions such as Discover and Prestige Financial Services to build machine learning models. Now DMS customers, including regional and national banks, auto lenders and property and liability insurance providers, can also tap the power of Zest AI and ML through procedures they already have in place with DMS. According to Dreux, this arrangement makes ML model building easier because DMS is doing the work of managing regulatory compliance and due diligence. On average, the Zest partnership should cut the time an existing DMS customer would need to develop an ML model independently by 75% to 90%.
“Zest is armed with a process and a solution for credit attributes. They have the tools to run the model. We have the tools to feed the model,” explains Dreux. “The two of us are building out a solution for an existing client base that is hungry. With Zest, we’re doing everything that fits with our mantra – connecting to data sources, making modeling faster and enhancing analytics.”