How We Use AI to Predict Mortgage Transactions

MonitorBase is a borrower analytics platform that lets you mine origination opportunities from your contact database.  Many of our clients ask for specifics of how we determine who in your database is likely in the market for a mortgage in the near future.  The short answer is we use Artificial Intelligence (AI).  MonitorBase has been using AI since before it was cool (2007).  Since AI is a broad term, I’ll give examples of how it applies specifically to mortgage origination and our users.  

ar·ti·fi·cial in·tel·li·gence
noun
the theory and development of computer systems able to perform tasks that normally require human intelligence, such as speech recognition or decision-making.  

The decision-making part of the definition is key.  For example, a lender determines who, of all the contacts in their CRM, they should reach out to today. That means who in today’s market conditions, is likely in the market for a mortgage within the next 30-90 days.  In the past, this was done by a human (Loan Originator) periodically by deciding who to send an email marketing campaign to.  Usually, lenders don’t base this on any data that would tell you that the borrower is in the market for a mortgage or even that they can credit qualify for a loan. Instead, they take the shotgun approach, sending them drip email marketing campaigns often based on arbitrary time periods or even worse, the dreaded birthday card. 

At MonitorBase we promote a culture of data-driven decisions. We have the expertise to monitor all of your contact’s underlying financial situations, lifestyle changes, and other non-mortgage behaviors that we’ve found often lead to a home purchase or refinance.  This data-driven approach is more effective because it will be hitting your contact’s inbox when they are in the mortgage or real estate mindset and are already credit prequalified for your products.  Is it possible for a human to keep track of the underlying loan terms, life changes, credit score, and propensity to buy a new home of the 500 to 1000 people they have a connection with?  Not likely.  That is where predictive analytics or AI comes in.  

How do we decide what behavior is predictive?

Social media platforms use something called a “look-alike” model to assist companies running advertisements to display their ads to people that have similar interests and behavior to the customers that have recently purchased their products. MonitorBase utilizes a similar approach by analyzing hundreds of data points on each individual borrower that just started the mortgage process and then finding who else in your database of contacts is displaying similar behavior at that same point in time.  Furthermore, as the real estate and interest rate markets continually change, so do our predictive algorithms. They are continually updated to reflect the current market conditions to ensure we’re identifying the right consumer behavior in the right market before alerting you.  Going back to the definition of Artificial Intelligence,  MonitorBase automates the human-like decision-making process of who to engage with each day, making you a super-human loan originator.   Click here for a detailed breakdown on Predictive Scenario Alerts.