Back to Industry Trends & Insights

The Key to a Quick Closing? Finding Errors Early.

Image Source: Shutterstock

Learn how cognitive robots can transform the way your company handles documents from start to finish and transform your process.

Alex Levi 
Vice President and Co-Founder

Remember that old saying about ‘the best laid plans?’ How they can (and almost certainly will) go awry? That’s the frustration which LO’s and Processors deal with every day, working to get the right and complete set of documents from their Borrower.

Although unintended reviewing those documents thoroughly before sending them forward to Underwriting can get pushed to the bottom of the priority list. With the phone ringing, emails coming and texts needing an answer, the additional deep look at the documents provided by the Borrower don’t always get the attention that they should. When they finally reach the Underwriters desk, wait to be reviewed and then are found to be missing necessary items, a week or more has often passed; causing detrimental consequences.

No matter how skilled and efficient your back-office is, if the paperwork they’re getting from the client is incomplete or inaccurate, then it makes a quick turnaround time very hard to achieve. 

In an industry where the speed with which you can process an application is key to your continued success, you need to find a way to get on top of your own paperwork and make it as easy as possible for your clients in the process. Cognitive robots and powerful AI software are making that a reality for companies willing to embrace the digital economy.

The Endless Hassle of Paperwork

Most people are careless with their paperwork; storing it in places they promptly forget about, using email attachments as a ‘filing system’, and then losing them or simply never filling in the docs they were supposed to complete.

It’s understandable. We all lead busy lives and looking after your various documents is always right at the bottom of the to-do list.

But the problem comes when you need your documents to achieve something, and you realize how dependent you are on them.

Dealing with The Broad Spectrum of Documents

Mortgage applications require a broad spectrum of documents that present a complete picture of who you are and your financial status, and it takes time and energy to pull that all together if you’re an average person with average paperwork.

Our job as lenders should be to make the process as easy as possible.

AI Foundry has developed cutting-edge technology that is able to recognize hundreds of document variations, such as

  • Bank statements, pay stubs, W-2s
  • Bank/mortgage/pension statements
  • Divorce decrees and escrow documents
  • Child support documentation
  • Property flood certificates and homeowner's insurance documents

With all of this documentation, there are bound to be errors such as missing fields, inaccuracies and other errors. AI Foundry’s cutting-edge technology has taken data gathering to a new level. Our platform automates processes by extracting data from electronic images and documents quickly with superior accuracy and minimal manual intervention. This is done by machine vision that extracts and classifies data and information and boasts a day-one accuracy rate of 99%.

Mistakes are inevitable when filling in a large amount of paperwork. The key is to pick up the errors quickly, make the applicants aware of them, and make it easy to fix. With a guaranteed 2-hour turnaround on any robot, and 4-hour turnaround on complete correspondent loan packages, our technology allows lenders to keep the data flowing through the system.

5 Steps in The Mortgage Application Process That Are Slow For Humans But Lightning Fast for Cognitive Robots

  1. Matching the borrower name on all documents - This is the kind of simple and vital task that is so tedious for people to do and so easy for cognitive robots. Finding fields on forms, checking spelling, sorting through piles to make sure no names or signatures are missing is light work with AI Foundry’s software.
  2. Validating social security numbers automatically - Let the system validate all the social security numbers that appear on documents. It’s a huge time saver for employees who normally have to do this step by hand and it’s far more accurately handled by robots.
  3. Matching the name and address of employers, past and present - many people do not realize that the name and address of all of their listed employers have to match exactly with the names and addresses they have provided in the loan application. It’s tedious and time-consuming for consultants to have to ‘stare-and-compare’ this data, while the robots can get it all done and compliant in a matter of seconds. 
  4. Checking the dates on the documents - documents that are older than 120 days are not admissible on a mortgage application. You should be running a system that can verify the dates on documents and instantly send notifications back to the client so they can rectify it. Those kinds of delays are frustrating for all parties involved in the transaction.
  5. Identifying an excessive number of properties - It’s so important to get this done before the loan gets into production. Putting in all that work on an application, and ordering third party services, only to realize later that the application will never be considered because the applicant has too many financed properties is crushing. Let AI take care of that automatically and save a huge amount of time. 

While AI Foundry has developed a wide range of cognitive robots, there are three specific solutions which have the biggest impact when it comes to gathering documents and preparing them for the next stage of the process.  

  • Document Indexing & Filing
    1. This cognitive robot automates the process of continuously monitoring and filing loan documents into the appropriate folders in the loan origination system (LOS).
  • Application Checklist
    • This cognitive robot automates the process of verifying all the application documents have been received for loans delivered to Fannie Mae.
    • Automatically creates a list of documents that are missing.
  • General Program Eligibility
    • This cognitive robot automates the process of verifying standard eligibility requirements for delivery of loan to Fannie Mae.

One of the most frequent problems companies face is the silo-ing of information between different departments. When it comes to gathering and processing documents, you need each and every team member to understand the flow and the process, which works better when the separation between back and front offices is collapsed.

‘Streamlined processing is critical for any successful lending business’, say the UX specialists Lightico. ‘If your organization hasn’t yet digitized their workflows and application processing, it’s time to consider how such a change might benefit your lending operations, improving the efficiency, time-to-close and ROI of loan application processing.’

At AI Foundry, we know that every organization is at a different point of their digital transformation journey. That’s why we collaborate extensively with our clients to build new systems to solve specific challenges or modify and streamline your existing business processes to make them more responsive, agile and faster.

Find out more about our cognitive robots, or simply request a demonstration to see what AI Foundry’s award-winning technology solutions can do for you.