As published in Scotsman Guide's Residential Edition, December 2008.
Many mortgage brokers’ businesses rely on the leads they purchase. They turn to various lead-providers hoping to receive strong prospects that will become funded loans.
Increasingly, lead-providers are scoring their leads to predict their likelihood of converting into funded loans or progressing to the next step in the loan-origination process. The concept is similar to how credit scores predict the likelihood of borrowers repaying a loan.
Brokers who purchase their leads from Internet lead-providers should understand how the lead-scoring process works, as well as how they can profit from using scored leads.
To determine lead scores, lead-providers analyze how data characteristics have influenced historical lead results. This is not complex when there is only one data field to analyze.
Subscribe to e-newsletter ___________________________
Want to stay up-to-date on news and notes from Scotsman Guide? Subscribe to our free, monthly e-newsletter.
For example, analysis may indicate that 50 percent of e-mail addresses from a certain domain will convert into funded loans compared to only 25 percent of e-mails from another domain. The lead score of the first domain’s e-mail leads would be 50, and the other’s would be 25.
The process becomes more complicated as we add more data fields, such as state and gender.
We may see that 60 percent of leads by males convert into a funded loan, compared to 40 percent for females.
And looking at the state field, we may find that Alaska has 70 percent of leads converting into a funded loan compared to 20 percent of Florida’s leads.
To determine a single, composite lead score, each field needs its own “weighting.”
In other words, it’s possible that state is a more accurate field to determine the likelihood of a funded loan than the e-mail domain, or vice versa.
Weighting each field’s degree of influence is the best method for normalizing these variations of influence.
From a data perspective, no two Internet leads are alike. Thus, the more accurate the score for predicting outcomes, the more valuable the process becomes.
Page: 1 2 3 Next