New tech trends have transformed the business of lending. Learn how leading lenders are using big data and predictive models to slash risk and boost revenue in our free webinar.
Lending in the personal loan space has traditionally been a very forward-looking leap of faith, where the lender does their best to create some sort of mathematical estimation of how likely it is that a borrower will pay them back, and then sets the price and hopes for the best.
And yet hoping for the best is not exactly a sound financial decision. “In competitive times, people get ahead of themselves in the interest rates they offer,” says Scott Crawford, VP of Product and Marketing at Ascend Consumer Financial. “And you can lose a lot of money.”
Classic underwriting practices can only go so far. Ascend is leaping to the head of the pack by looking at data in innovative new ways. Their proprietary analytic model, called Adaptive Risk Pricing, is designed to create a closer alignment between risk and pricing. It goes beyond purely predictive models to incorporate dynamic pricing into consumer loans by identifying new sources of data and new ways to identify material, valuable correlations — and that’s the key.
In today’s Big Data approach, Crawford says, when you’re looking for trends, there’s a tendency to throw in everything but the kitchen sink. That can be powerful, but there are dangers with that approach. “You want to make sure you can bring everything back to some sort of theoretical underpinning that actually makes sense,” he adds.
Those are the kinds of checks that will ensure that your model actually performs well if conditions change. “The trouble you have with an overfitted model,” he continues, “is that if conditions change your model is going to blow up because it doesn’t reflect reality anymore. If your model is based on sound theory, then it has a better chance of holding up.”
For Ascend, it’s real-time data: current financial situation, cash flow, and smart choices. These are the three Cs of classic risk models — capacity, character and collateral — uncovered in innovative new ways and from a wide variety of sources that lenders have not traditionally incorporated and used to build out new models.
It’s a massive problem in the near-prime space, he says — the lack of visibility into a borrower’s true financial situation and a lot of really uncertain financial information.
“That’s where we think the biggest gap is in the traditional underwriting model,” Crawford says, “and that’s where we focus on filling it.”
To learn more about how tech, data, and predictive modeling is transforming lending, take an hour and tune in to our panel of pros — questions welcomed!
Don’t miss out!
In this webinar you’ll:
Learn the trends driving change in marketplace lending and credit risk management
Build more accurate predictive ratings models based on alternative data sets
Get insight into the future of marketplace lending and credit risk management
Scott Crawford, VP of Product and Marketing, Ascend Consumer Financial
Krishna Venkatraman, SVP of Data and Analytics, OnDeck
Terrence McKeown, Practice Manager, Credit Analytics, Envestnet | Yodlee
Evan Schuman, moderator, VentureBeat
This webinar is sponsored by Yodlee.