Mortgage Closings in 15 Minutes with Snapdocs on AWS

2020

If you've recently closed on a mortgage, you may have been struck by how paper-based the process can still be. Although many aspects of our financial lives are now handled digitally, buying or refinancing a home can require reviewing, initialing, and signing dozens of documents from multiple parties during face-to-face meetings that can last hours. This opaque, manual, and fragmented process can leave homebuyers confused and stressed.

These are some of the problems Snapdocs wants to solve. Snapdocs is a US-based real estate technology company that was founded in 2014 with the goal of streamlining the mortgage process with innovations based on artificial intelligence, machine learning, and other cloud computing technologies.

The company first used Amazon Web Services (AWS) technologies to build tools that enabled title companies to schedule and manage notaries for mortgage-signing appointments nationwide. Since then, Snapdocs has expanded its AI capabilities to enable automation, accuracy checks, and security controls throughout the mortgage-closing process. The latest Snapdocs release is a suite of machine learning solutions, also running on AWS, that enable a digital mortgage-closing experience for mortgage lenders, title companies, signing agents, and homebuyers at unlimited scale.

"The traditional mortgage-closing process can take as long as 60 days and cost upwards of $6,000—and that's only if no errors are discovered along the way," says Briana Whelan, head of product for Snapdocs. "We saw an opportunity to use AWS to build a solution that could help lenders and title companies both operate more efficiently and provide the digital closing experience their customers are increasingly demanding. The Snapdocs platform now processes one million transactions annually, a scale that never would have been possible without AWS."

start a python tutorial
kr_quotemark

With a platform underpinned by AWS, we've been able to reduce errors in the closing packets by an average of 75 percent."

Briana Whelan,
Head of Product, Snapdocs

Simpler Document Digitization with Machine Learning on AWS

Even before Snapdocs launched this platform, some mortgage closings were being transacted digitally, although not many. "The problem was that digitizing real estate transactions was generating more work for lenders and title companies, not less—the opposite of what most people expect from going digital," says Mike Phillippi, head of marketing at Snapdocs. "It sounds simple to digitally sign documents, but the mortgage industry operates under hugely varying regulations from jurisdiction to jurisdiction across the country. And the companies need to figure out how to integrate with hundreds of others that have very different levels of technology adoption. All that is just too heavy a lift for any single lender or title company to take on at any kind of scale."

That's where Snapdocs comes in. The Snapdocs digital closing platform is accessed by lenders, title companies, and signing agents through a web portal, agnostic of the IT infrastructure of the connecting company. Snapdocs AI automatically ingests closing documents regardless of what technology a lender is using to generate them, eliminating the time needed for manually transferring documents from participant to participant. Additional AI bots then go to work, classifying and presenting documents in the right order, annotating where signatures and initials are required, and verifying that every signature is where it’s supposed to be.

Snapdocs customers benefit not just from a simplified document-digitization process but also from the ability to exchange those documents through a platform that preserves and communicates important contextual information. "There's efficiency to be gained just through effective, value-added document digitization," says Whelan. "Because Snapdocs enables exchange through a platform, our customers also benefit by having insight into where documents have been, where they're headed next, and what parts of the packet require their attention now."

Snapdocs created this platform using proprietary machine learning models that it trains on Amazon Elastic Compute Cloud (Amazon EC2) T2 instances powered by Intel Xeon processors. To deploy these models, Snapdocs uses AWS Lambda, which runs serverless code in response to events; Amazon API Gateway, which can accept and process hundreds of thousands of concurrent API calls; and Amazon SageMaker, a managed end-to-end machine learning service that makes the models available for inference processes at HTTPS endpoints.

Improved Efficiency and a Better Customer Experience

The flexibility of AWS technologies helps Snapdocs provide the customized experience that is essential in an industry where there are almost no universal processes and workflows. "Running on AWS enables us to automate the overall process as well as scale for and incorporate all the variations needed across partners and jurisdictions," says Phillippi. "In the past, borrowers were looking at 1–2 hours going through confusing legal documents. Now, with Snapdocs on AWS, that final meeting can be as short as 15 minutes."

By standardizing on AWS for its cloud computing and machine learning platform, Snapdocs has been able to automate error detection and greatly reduce the expense and inconvenience such errors can cause. "With a platform underpinned by AWS, we've been able to reduce errors in the closing packets by an average of 75 percent," says Whelan. "When errors aren't detected until everyone's at closing, you have to cancel and start the process over. It's costly for lenders and title companies, and it's a huge hassle for the homebuyer."

Because of the scalability available through AWS, Snapdocs can serve any mortgage lender, title company, and signing agent in the United States without drastic infrastructure challenges or needing extended human resources for manual work. "We now touch about 11 percent of US residential mortgages and are growing fast," says Whelan. "In the past, processing that level of volume across numerous lenders would have required unsustainable manual work. By using a machine learning solution built on AWS, we not only avoid that manual work, we also make the whole process easier, faster, and less error-prone for everyone."

Most important, Snapdocs is turning the promise of digital closings into a reality for its customers, most of whom couldn’t have devoted the time and attention needed to achieve this capability on their own. "Previously, for lenders that wanted to offer digital closings, there was a tension between their own efficiency and their customers' experience," says Whelan. "With Snapdocs on AWS, they no longer have to choose one or the other. Our platform helps them streamline their internal processes at the same time that they provide a high-quality digital experience for their customers."

To learn more, visit aws.amazon.com/intel.


About Snapdocs

Snapdocs is a real estate technology company that uses integration and automation tools to streamline the mortgage closing process. The company’s products enable uniformity, transparency, and compliance across thousands of jurisdictions.

About Intel

Intel is the world’s leading designer and manufacturer of high-performance processors for servers, PCs, IoT devices, and mobile devices. AWS and Intel engineers have worked together for more than 10 years, building custom hardware to ensure AWS services run on a platform optimized for customer workloads for the best value. Intel Xeon Scalable processors power Amazon EC2 instances to help enterprises drive performance for their compute-intensive workloads.

Benefits of AWS

  • Reduces real estate closings from 1 hour to 15 minutes
  • Eliminates 75% of errors
  • Improves efficiency and customer experience
  • Supports compliance in thousands of local jurisdictions
  • Preserves valuable contextual information

AWS Services Used

Amazon Elastic Compute Cloud

Amazon Elastic Compute Cloud (Amazon EC2) is a web service that provides secure, resizable compute capacity in the cloud. It is designed to make web-scale cloud computing easier for developers. Amazon EC2’s simple web service interface allows you to obtain and configure capacity with minimal friction.

Learn more »

AWS Lambda

AWS Lambda lets you run code without provisioning or managing servers. You pay only for the compute time you consume.

Learn more »

Amazon API Gateway

Amazon API Gateway is a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure APIs at any scale. APIs act as the "front door" for applications to access data, business logic, or functionality from your backend services.

Learn more »

Amazon SageMaker

Amazon SageMaker is a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly. SageMaker removes the heavy lifting from each step of the machine learning process to make it easier to develop high quality models.

Learn more »


Get Started

Companies of all sizes across all industries are transforming their businesses every day using AWS. Contact our experts and start your own AWS Cloud journey today.