Tyler Sawyer: Lenders Must Data Model the ‘Most Likely Borrower’


PERSON OF THE WEEK: Automation of the mortgage application process has become increasingly sophisticated. Lenders today are almost universally deploying advanced online portals where borrowers are walked through the mortgage application process, whilst on the back end automated verification of income and assets and automated underwriting make the process smooth and seamless, with every step of the process recorded along the way.

More fancy rigs make use of artificial intelligence, or AI, to further streamline the process and wow consumers with a “virtual concierge”-like experience.

But a question remains as to how lenders will drive qualified leads to these fancy portals. What strategies will work best?

Perhaps the answer lies in using data to determine who is most likely to buy and borrow.

Tyler Sawyer, vice president of rental and real estate for Equifax, recently discussed the important role technology is playing in capturing qualified leads in this current, highly competitive mortgage lender market.

Q: What should lenders do to remain competitive amid the current compressed market?

Sawyer: To remain competitive, lenders must better-navigate between those ready to borrow and those who are only interested in one-day borrowing to buy.

Interest in homeownership is strong right now. Bank of America’s most recent 2018 Homebuyer Insights Report indicates 72% of millennials prioritize owning a home, yet only 37% are homeowners.

Likewise, many homeowners are electing to “love the one they are with” and remain in their current home with the low interest mortgage they secured years ago.

Some of these homeowners are leveraging products like Home Equity Lines of Credit (HELOCS) for renovation/expansion projects vs. upgrading to a new home – and taking on a more expensive loan.

This disconnect between “potentially interested” consumers and truly active prospective borrowers is what lenders must successfully navigate.

These factors are forcing lenders to become more proactive in identifying borrowers who may be looking for a new home (or more importantly, about to look for a new home), or those looking to finance the renovation of their existing one. 

Q: How critical is it for lenders to become more proactive in order to engage borrowers? 

Sawyer: Lenders that simply cold call consumers at random to generate leads are likely too see poor results because the pool of borrowers requiring immediate mortgage services is significantly smaller than it has been historically.

To find success in this new market environment, successful lenders will leverage technology that models the likelihood that a potential borrower is in the market for a first-mortgage, refinance and/or HELOC.

Proper modeling employs robust data analytics to better identify and narrow those who are more likely to borrow in the near term, which helps lenders better prioritize their resources to actively engage with qualified prospective borrowers today and follow-up with those leads that may need additional time to mature later.

Q: What data do the models utilize to identify the prospective borrower most likely to move forward to a lending process?

Sawyer: Proper modeling aggregates data down to a trend-specific level. In this process, patterns within a particular geographic area are examined and compounded with wealth and asset information for further analysis. This data is then studied at the property level.

After the models are generated, leads are vetted using a unique customer score. These scores range between the top 10, 15 and 20 percent of borrowers who are more likely to express interest in the modeled loan product. If the best data is used properly, lenders can realize a 300 percent increase in qualified mortgage applicants versus typical results in marketing without data – including the cold-calling of existing borrowers who a loan officer believes may be ready to move, refinance or renovate.

Q: How is borrower data aggregated in order to remain compliant while also being used for modeling?

Sawyer: Given regulatory restrictions that may be involved in dealing with consumer credit data, it’s important to interact with the consumer data at a high level. The models aren’t looking to dive deep into an individual’s information, rather leverage the data to analyze trends seen among similar consumers to help borrowers gain a better idea of who the market to.

To meet that objective, the modeling process anonymizes and aggregates sensitive data, yet enables lenders to better tailor their outreach to market more effectively to a specific demographic.

Q: What results should lenders expect to see?

Sawyer: As mentioned, when the right data is used, we are seeing lenders increase qualified applicants three-fold. Leveraging better data analytics helps lenders market their products more effectively and, ultimately, increase sales.

Look at the industry’s excitement around creating a frictionless, streamlined approach for the borrower experience. That is great, but the conversation is almost nonexistent when it comes to attracting borrowers in the first place.

If mortgage lenders fail to identify the best qualified borrowers through better data analysis, they will lose the opportunity to deliver the improved borrower experience that consumers expect and in which they have so heavily invested.

Notify of
Inline Feedbacks
View all comments