Data Drilling, Search Engines Can Be Problem for Field Servicers

Mortgage & Real Estate

Data analytics has been universally embraced by the mortgage industry but contrary to popular belief many field services companies are missing out on their business potential, says Tim Grace, CEO of PointPredictive.

Various providers are using technology to amass mortgage fraud, home values and rental market databases. Nonetheless, property preservation is one of the mortgage industry areas “that are currently underserved with analytics,” he says.

Today too many companies “are building complex algorithms that business users don’t understand and therefore don’t use effectively,” according to Grace, who claims that everyday experience “tells that now the market needs a new breed of analytic models that are simpler and more intuitive for the end user.”

Banks, lenders, servicers and property management firms “have no real analytic tools to help them differentiate how to take care of propertiesbased on the house itself, the neighborhood, or crime rates,” he explained. The best way to take care of each property in a more cost-effective way is by collecting data and creatively using scientific methods that target existing and potential problems.

The PointPredictive approach is based on “gathering data in consortia, sharing information between the relevant parties to build analytics and predictive models” that show consortia and data sharing are a very efficient tool, Grace adds. The firm is working to help facilitate that approach in the mortgage industry and property preservation management “for many different types of business challenges.”

Tighter oversight and regulatory restrictions on how mortgage servicers collaborate with the third-party field services industry also are challenging the industry to improve data drilling and analytics.

At the MBA’s National Mortgage Servicing Conference and Expo 2014 in Orlando a session on emerging technology and data analysis strategies and tactics for property preservation and bank owned foreclosures is just one of a series of upcoming discussions that will inquire about solutions provided by specialized mortgage technology vendors in business intelligence and analytics, mobile options in the field, and cloud computing in the property preservation market.

Safeguard chief information officer George Mehok, who is one of the panel speakers, says he will introduce two of the firm’s most used technologies: the MapAlert weather assessment service and a mobile technology for vendors that rely heavily on data analytics that generate business intelligence. The MapAlert system utilizes internal data, geo-mapping of 15 million properties, information about severe weather events from several sources, and Google mapping technologies to help determine the potential for property damage.

Mehok finds data analytics as a part of property preservation services has to continuously evolve. “It is essential to demonstrate compliance with regulatory requirements” to generate internal performance evaluations, scorecards and dashboards, evaluate vendor management network, to assess their performance, allocate work orders and identify market needs.

Data are very important, but this is not about the math, says Keven Smith, president and CEO of Mortgage Builder, and a mortgage servicer’s experience is what determines the focus and success of data drilling and analytics.

Servicers and vendors who really know the software and the data they have within each servicing system, work with computer programmers to link it with the databases and systems that each mortgage company uses to manage loan data.

Data analytics has always been there, he adds, but most recently the focus has moved on. “How do we go in and build something more dynamic to allow users maximize the value of data? It’s not a math problem, it’s more of a mortgage servicing issue of what lenders really want to see and what’s important data-wise and how to make it user-friendly interface for lending mangers,” so they can determine how to drill and analyze loan and property data, and what results they want to see. For example, loan types and how they are being serviced.

In other words, computers and software power is determined by its user’s expertise, the deeper the market expertise these users have, the better systems they will be able to help generate. Ultimately, data drilling strategies become even more important than the data itself.

“Data analytics is much more important today than in the past, and will only become more prevalent as time goes by,” Smith says.

“The mortgage industry has faced has forced managers were forced to look at data more seriously, which is a good thing because now people understand their businesses much better and that will only make the industry stronger.”

The vision has to constantly change alongside the need for data analytics, agrees Grace. The key is in breaking out of predictable operations.

He maintains “there is an art and science to building great predictive models.”

PointPredictive analytics are designed “to bridge the gap between sophisticated science and the practical things that business users need those algorithms to accomplish.” Staff includes a group of math wizards who can help transform business experience into effective technology developing “hundreds of analytic models across multiple industries in the past two decades,” such as debit card fraud prevention software for MasterCard a few years ago, says Grace.  In the past few years the team has been focusing on deployment technology solutions for the mortgage industry that allows faster deployment, easy data acquisition and access.

“In practical terms, data is not always clean or complete,” Grace says, which makes it necessary “to effectively work around these realities and get creative.” Mortgage companies constrained by internal resources, “such as lack of domain expertise, system integration constraints,” or other challenges, need a targeted way to generate more cost-effective algorithms.

Leave a Reply