Getty Images/iStockphoto

Mortgage data vendor uses Qlik to build analytics platform

Polygon Research used tools from the analytics vendor to develop a SaaS platform consisting of nine dashboards that mortgage specialists can use to inform decisions.

In need of a way to get pertinent data in the hands of decision-makers throughout the mortgage industry, Polygon Research used embedded analytics capabilities from Qlik to build a series of SaaS applications.

Founded in 2015 and based in Washington, D.C., Polygon Research is an independent software vendor that develops data products for the mortgage business. The vendor collects data disclosed by lenders and regulatory agencies, integrates and prepares it before plugging it into data science models, and develops analytics tools including reports and dashboards that clients use to inform decisions.

Qlik, meanwhile, is a full-featured analytics vendor whose tools can be used throughout the analytics pipeline, from data integration through analysis and insight, and are aimed at resulting in data-informed actions. Most recently, the vendor launched Qlik Cloud Data Integration, a cloud-based data integration platform as a service.

The problem

The mortgage industry depends on data to drive key decisions.

Each loan applicant carries inherent risk of default, so lenders use data to carefully examine whether a given applicant is likely enough to pay their loan to make it worth the risk of lending money to that applicant. In addition, both the secondary mortgage market, where loan servicing rights are bought and sold, and the loan securitization process, where loans are packaged and traded as securities, depend heavily on data.

Mortgage data, however, is voluminous and isolated in spread-out data sources, with loan originations in the United States totaling $2.7 trillion in 2021, according to the Federal National Mortgage Association -- commonly known as Fannie Mae -- and more than 4,500 lenders nationwide that don't share data with one another.

Data includes key information about regulations, borrowers, real estate brokers and agencies, and properties.

Screenshot of Greg Oliven, CTO of Polygon Research, on video during a webinar.
Polygon Research CTO Greg Oliven speaks during a webinar hosted by Qlik.

"The mortgage industry is quite large, and there's a lot of value at stake," said Greg Oliven, CTO at Polygon Research, on Nov. 30 during a webinar hosted by Qlik. "The industry's data is large and unwieldy, which we call the last-mile problem, which is how end users unlock the value contained in [the industry's] data sets."

So, in 2018, Polygon Research set out to build a set of applications that unlock the value contained in the mortgage industry's massive data sets and make mortgage-related decision-making more informed.

Lenders, servicers and investors had their own historical information to inform decisions, but if they wanted broader context, they had to comb through external data themselves and develop data products on their own.

Some larger organizations might have had the resources to do so, but that wasn't the case for all mortgage-related enterprises. And even for those that did have the resources -- including data scientists and financial capital -- collecting data from disparate sources, modeling the data and developing analytics tools from the data is a time-consuming process.

We've set out to solve the problem of unlocking the value contained in industry data sets.
Greg OlivenCTO, Polygon Research

Polygon Research, therefore, aimed to remove the burdens of data management and data preparation from individual mortgage enterprises and provide them with data products that lead to insights.

"We've set out to solve the problem of unlocking the value contained in industry data sets," Oliven said.

The answer

To address the needs of mortgage lenders and investors, Polygon Research used Qlik's analytics platform to build the Polygon Research Platform, which consists of a series of applications replete with embedded data that customers can subscribe to as a service and use to inform decisions.

Initially, Polygon Research chose to use tools from a vendor other than Qlik, though Oliven declined to name the analytics provider. Qlik has numerous competitors, including embedded analytics specialists like Logi and Sisense, data integration vendors including Informatica and Talend, and full-featured analytics vendors such as MicroStrategy, SAS and Tableau.

According to Oliven, the platform that Polygon Research initially chose couldn't scale to meet the vendor's needs. So Polygon Research instead turned to Qlik. And using Qlik, the vendor built the Polygon Research Platform.

"We switched horses and never looked back," Oliven said. "We went with Qlik because of the scripting within Qlik Sense, and the scalability and cloud story."

Qlik tools that Polygon Research used to develop its mortgage-specific analytics platform include the following:

  • the Qlik Associative Engine as the base for building applications with multiple use cases such as sales, compliance and automated machine learning;
  • Qlik's in-memory model to enable speed despite the large amount of data contained in each application;
  • Qlik Cloud to build a scalable platform that can handle even more data as it comes in from existing sources and new ones that might arise;
  • Qlik's automation engine to create scheduled reports and update data sets with new information; and
  • low-code/no-code capabilities to enable easy development by Polygon's developers and end users to easily explore the data in each application.

"[Qlik] has a powerful interface that we really like," Oliven said. "It's a 'write once and leverage many' kind of environment."

He added that Polygon customers do not have to be Qlik customers to consume the information in the Polygon Research Platform. But if they are joint customers of both Polygon and Qlik, it enhances their experience.

"We are able to empower our clients through the interface without them needing to be a Qlik client, but it's always great if they are [a Qlik client] because they can leverage their understanding and investment in the environment," Oliven said.


Ultimately, Polygon Research used analytics capabilities from Qlik to build nine interactive dashboards -- each embedded with numerous data sets that automatically update -- making up the Polygon Research Platform.

Customers can subscribe to as few as one of the nine dashboards, some combination of them or all of them. Pricing for Polygon's various dashboards ranges from $1,499 annually for CensusVision and FHAVision to $7,299 per year for SPCP (Special Purpose Credit Program) Analytics-in-a-Box.

The nine dashboards Polygon Research developed are designed to give customers what the vendor calls an integrated framework informed by data.

They include data sets that enable users to compete with peers at the local level, including the development of a sales strategy; identify new markets and segments within markets to spur growth; implement diversity and inclusion goals in both hiring and customer acquisition; and track both internal KPIs and external market performance.

Within each dashboard, customers can explore the data sets without having to write code.

"They can interact with microdata, really drill down and look at a particular market or a loan product," Oliven said during a demonstration of HMDAVision, which costs $4,000 annually and enables users to look at specific geographic markets. "It works at a very granular level. And you're able to pack a lot of information into one interface."

Dig Deeper on Business intelligence technology

Data Management
Content Management