Mortgage rates set the 14th record low of 2020, causing a surge in mortgage refinancing.
The crush of loan applications that swamped the mortgage industry in 2020 will not abate anytime soon, according to these recent headlines. Bankrate states that “record low rates may also usher in 2021.”
Historically low mortgage rates, combined with the pandemic’s ongoing pressure on household finances, are driving more borrowers to refinance their home loans. And with the new vaccines rolling out across the USA in 2021, we can expect an explosion of new home purchases as the pent-up buyer demand is released, and more sellers have confidence to put their homes on the market. “Without a doubt, there are too many buyers in relation to the sellers,” said Lawrence Yun, chief economist of the National Association of Realtors.
Must Go Faster
In a new report, Infosys has identified the top trends for mortgage processing in the post-pandemic era. Trend number 5 is accelerating the mortgage loan onboarding process.
The shift away from paper documentation toward a complete digital loan origination is projected to take years for the mortgage industry. This is because the industry does not use a common platform shared by all the players involved in the process to avoid duplication and repetitive tasks.
Lack of Standardization
This failure of standardization results in gaps between aggregators’ point of sale platforms and mortgage lender loan origination system platforms because information is not often seamlessly transferred between these systems. As a result, any information passed on, such as the presence of a document type and its contents, must be revalidated and confirmed. Due to the complexity of the documents and processes involved, over 60% of the time is spent on manual operations.
Due to the complexity of the documents and processes involved in mortgage loan processing, over 60% of the time is spent on manual operations.
A single mortgage loan file can contain over 100 diverse loan forms and documents. Federal agencies, states, counties and cities have different forms and regulations with an endless variety of file formats. Applicants cannot be counted on to submit high-quality scanned or phone camera snapped documents. Poor quality source documents are the curse of every automated system. Loan files often contain images that are unreadable by most systems. Applicants also forget to file documents.
Loan File Review
Loan file review is the most critical stage of the loan process. The steps involve reading the file contents, sorting and verifying the documents, and extracting data. It is also an incredibly labor-intensive process prone to human and software errors. If anything goes wrong at this stage, the rest of your process is affected.
As a mortgage processor or service provider, you must provide a speedy customer experience for two fundamental business reasons:
- Faster processing means faster capital realization for your company.
- Delays mean you risk losing your customers to competitors.
Smart OCR to the Rescue
Infosys is recommending that mortgage lenders invest in smart OCR powered by AI/ML to mitigate these challenges and accelerate the loan onboarding process. The report cites several benefits of deploying a smart OCR solution:
- Validate immediately the documents submitted by the borrowers and the complete loan package shared by the correspondent lender or mortgage servicer. This instant validation allows the lender to quickly request any incomplete or missing information.
- Verify the presence or absence of all the mandatory documents. This alone can save lots of downstream process time.
- Extract the relevant information from documents, validate the data and pass it on to downstream applications without any delay.
- Sort all the document versions from initial to final. In OCR terminology, this is referred to as document classification.
- Reduce the processing time by up to 50% and reduce the cost of quality control and human effort.
The Infosys report also recognizes that the level of automation possible is directly correlated to the intelligence and accuracy of the OCR solution.
The problem is that most existing solutions are based on legacy OCR technology that hasn’t kept up with modern advancements such as deep learning and therefore isn’t “smart”. This means a tremendous amount of work is required to configure and maintain systems that can at best achieve relatively low levels of automation.
When changes are needed due to new business or regulatory requirements, it often takes a significant amount of effort to update these legacy systems. If you need to process information on supporting documents such as detection and verification of notary stamps, signatures and initials, the old solutions require too much manual intervention. Against a backdrop of increasing customer demand with expectations of quicker response times and processes, the problem just won’t go away.
Get Smart with Parascript
Parascript Smart Learning is a document-focused learning platform that combines decades of document and image expertise with the latest in machine learning technologies. For mortgage loan processing, Parascript can deliver high levels of reliable automation without the need to write a single line of code and without the need to have trained staff.
Discreet machine learning algorithms, tuned to solve specific tasks, are expertly applied to solve the widest range of automation tasks. The result is a significantly faster time to get the system up and running in production. The system is capable of auto-configuration and can run unattended without the need for supervision.
Parascript customers report up to 50% more mortgage document automation with seven times (7x) the throughput compared to a competitive solution. Parascript achieves this level of automation at about 1/100th of the cost and effort.
Parascript Smart Learning is also the key to reaching the highest levels of straight through processing. This means that more data is processed by the software and never has to be reviewed by humans. The system can intelligently discern between good data and data that genuinely requires review. The results are improved throughput and lower costs, with greater adaptability, accuracy and certainty.
About the Author
Dan Lucarini is an AIIM Fellow with over 25 years’ experience in capture and content management technologies. Learn more at www.linkedin.com/in/danlucarini.