The Science of Lending Leads: Mining Alternative Data in a New Mortgage Marketing Universe

By Clint Lotz

As the COVID-19 pandemic continues, lending institutions across the country are monitoring major lending trends to determine which types of loans are most in demand.  Amid the uncertainty of the current economic climate, there has been a consistent and unprecedented consumer demand for mortgage loans and refinancing, as a result of consumers refinancing their current properties, or city dwellers seeking refuge in the suburbs. New home sales are projected to have increased 14 percent in 2020 compared to 2019, according to a recent forecast from Fannie Mae.

This trend, in combination with historically low interest rates, means that lenders are projected to lend $3.9 trillion in 2020.

Though this may sound beneficial—increased inbound inquiries for loans with decreased, or even passive effort for lead generation—it’s actually problematic, as lenders are unable to utilize new credit data fast enough to predict the financial health of the U.S. population, thereby crippling their efforts to meet the onslaught of demand.

Mortgage and lending marketers are now forced to fight for every qualified lead, while working to maintain or gain a competitive edge in the marketplace. At a time when the loan process is being sped up to keep up with shifting demand, and the market is being flooded with applicants who don’t qualify because of their mediocre credit scores, lenders are seeking out the leads that will result in successful loans. Moving forward, lenders are now reevaluating their lead processes and their risk models.

Marketing costs remain consistently high

Though the lending landscape is changing in response to the pandemic, the cost of traditional marketing is still high for mortgage brokers, as the cost-per-click to find qualified leads averages around $90. Multiplied by the thousands of leads a lender is targeting at any given time, it’s easy to understand why marketing budgets are high. These costs only increase once lead aggregator sites (such as Lending Tree) are involved. Lead aggregator sites send leads to multiple lenders simultaneously, while bombarding customers with competing offers all at once. Though most lenders employ automated software that can process lending applications in mere seconds, the reality is that only a fraction of those applications (on average about 30 percent) of those applicants are viable.

In addition, in 2020 Google implemented new restrictions limiting ad copy pertaining to credit, qualifying and loans.

The combination of all these factors results in expensive cost of acquisition numbers, as lenders continue to reach out to consumers with healthy credit, thus continuing the need for expensive marketing outreach.

Impending credit havoc

The millions of forbearances and loan modifications that took place in the early months of the pandemic are going to wreak havoc on people’s credit reports in the near future. Many consumers have balloon payments coming due. Given the overall state of the economy, lenders can expect to see missed and late credit card, mortgage and car loan payments.

Overall, these COVID aftershocks will impact credit reports for some time to come and will make finding qualified applicants more difficult for bank marketers.

With the fallout of COVID-19 combined with the usual pain points to reach qualified consumers, what can marketers do to improve applicant qualifications while reducing customer acquisition costs?  The answer is for them to mine their own resources.

Dig deeper in the mine

A typical lender is likely sitting on hundreds of thousands of contacts, if not millions of people who may have applied for loan products in the past and did not qualify at the time. Rather than throwing money at new leads, the best bet for marketers is to mine their existing leads. Software that can activate their existing databases and data sets to uncover new prospects can help lenders find “hidden gems” who might already qualify or will have better credit in the future—providing lenders with prospects to retarget with customized loan offerings.

With a glut of potential leads to dig through, here are some useful steps for banks:

  • Explore fintech options. New fintech technologies are being developed, launched and proven in the marketplace daily. With the rise in data science and AI practices, it is easier than ever for lenders to build underwriting models in house or work with fintech companies to create them. When done correctly, these algorithms can sort through large amounts of data to best determine which prospects are credit worthy and which are high risk.
  • Implement automated segmentation strategies across CX and sales teams. By implementing alternative data and CRM tools, marketers can create signals to indicate potential opportunities and forecast the needs of their database contacts. With automation, this data and these tools can provide a visible pipeline to sales teams for follow up online and offline.
  • Take alternative data seriously. The old way of evaluating personal credit is not going to work in 2021. Risk models need to adapt, and marketing lenders need to use alternative data to manage this. FICO is now revising their scoring to weigh the impact a personal loan has on a credit score, as a result of the sheer popularity of personal loans—which have led to easy approvals, funding (thanks to the rise of fintech lenders), and too many consolidation loans. FICO is now using an alt data category called “trending data” to track (and score) how well a consumer manages debt levels across all accounts.
  • Employ a data science team. If this is within your company’s budget, create a data science team, or at least designate one person for this role. Many lenders are using large-scale business intelligence tools to provide clarity to their data-mining efforts, but these tools don’t provide strategy. A data science team can use these tools not only to analyze data, but they can then help determine best use cases to ultimately grow a lender’s business.
  • Consider new product creation. From refinancing to collections, the use of alt data can help create new products that reduce churn and improve CX.

By taking these steps, lenders can not only become less reliant on third-party lead aggregators, they can reduce their overall COA. At the same time, lending marketers can expand the amount of credit qualified approvals, retain existing customers longer and create customized loan offers based on the customer’s background and needs

Clint Lotz is the president and founder of, a company that specializes in predictive credit technology that helps determine consumer lending potential.