In this Experian Business Chat we turn our attention to the record flood of new businesses getting started and, the need for credit grantors to evaluate credit applicants with greater speed and precision. To help us understand how that job gets done, we invited Product Director Greg Carmean to join us. Greg leads product development related to our partnership with Small Business Financial Exchange, so we met to talk about our partnership, and how our focus on innovation is enabling emerging businesses to succeed while giving lenders greater ability to identify and respond to risk.
What follows is a lightly edited transcript of our discussion.
[Gary]: So Greg, with so many new businesses entering the market, one of the things that we hear often from clients is a need for a better view of their customers, to more accurately assess risk. How is Experian innovating for small business service lenders in that regard?
[Greg]: Well, Gary innovation starts with the best possible raw materials, and in our business that’s data. So it really starts with our focus on obtaining credible information on both businesses and business owners from Experian and then also supplementing that with information coming from the Small Business Financial Exchange, which is considered the gold standard in lender data. We take that data through a data quality process, which is part of the innovation. We call it our Data Foundry, which really makes sure that we’re operating in the cleanest version of the data we possibly can. And then we apply our analytical techniques to provide the most predictive outcomes for our clients so that they can create the most tailored treatment for their small business customers. Finally, we provide platforms that enable not only list distribution, especially as we become so much more focused on the digital economy since the onset of the pandemic, but also platforms that are analytic in nature so that our client’s data scientists can work with the data in formulating the best possible strategies for the business.
[Gary]: That’s great. So Greg, what can lenders do to improve decision-making across the customer life cycle and how important is it to see business credit risk over time and how can that be viewed by a lender?
[Greg]: Gary, what we found is that the trends in the data are tremendously predictive in the risk outcomes that our clients are trying to predict. We’ve known this for some time on the consumer side of our business, as we’ve had trended attributes there for many years. What we’ve done now with the SBFE program at Experian is we’ve developed our V3 attributes, which are trended, and this trend in nature, not only allows you to see the static view of where a business attribute is at a point in time but at how the business attribute got there. Is the business improving in credit quality, are they declining in credit quality? And that is tremendously powerful when you incorporate those types of attributes into scores.
[Gary]: So trended attributes, they sound powerful, can you discuss the use of some of the different types of attributes?
[Greg]: Sure. Gary. So in addition to the trended nature of our V3 attributes, we’ve also incorporated a variety of different specialty attributes. For instance, how do businesses perform on government funding programs like SBA loans and the PPP program? How can I understand how a business is going to operate under forbearance programs? And also how could I potentially do things like reject inferencing to see if I’m leaving accounts on the table. These are all elements that we’ve worked into this V3 version that can further help our clients refine their strategies and get the best business outcomes.
[Gary]: So how do lenders manage accounts once they’re found and, and handle regulatory changes across the account mix?
[Greg]: Well, Gary, one of the most impactful things that we’ve done through the SBFE program is really focused on what we call machine-learned models, and machine learning, I think most people are familiar with, it’s the notion that a model can continuously take in additional data and learn over time. But when it comes to regulatory compliance, that can be challenging because the model isn’t very stable. So what we’ve done is we’ve adopted a strategy where we use all the machine learning techniques right up until the time the model is deployed, at which time it becomes static. What that allows us to do is create a model that has, as an example, consistent adverse action codes that will come out of it, which is tremendously important for regulatory compliance. Another element is making sure that the model is transparent so that a regulator can understand at a human level, how we get from the input to the decision. With our approach to these machine-learned models, we’ve been able to accomplish that and the results have been tremendous for our clients.
[Gary]: So you lead product development related to our partnership with the Small Business Financial Exchange. Can you speak to how these solutions benefit SBFE members and empower small businesses?
[Greg]: Absolutely. Both SBFE and Experian are committed to empowering small business growth. And we do that by placing powerful tools in the hands of our clients who are SBFE members so that they can put the small business that they’re dealing with into the right product at the right terms. The second thing is as they are managing those accounts, our tools really help them identify risk earlier, which allows them to intervene more proactively. And that can really make the difference in a small business account that can be maintained versus one that might go to charge off.
[Gary]: Well, Greg, this has been very insightful. It sounds like we’ve got a lot going on to benefit both small business service lenders and the businesses they’re serving. Thanks for coming on and sharing your insights.
Take the next step
If this chat raised your curiosity about Small Business Financial Exchange attributes for small business lending, consider watching the recording of our webinar titled “Disruptive Strategies to Empower Small Business Lending”
