.png)
How can OCR be used to improve Mortgage Applications?
First, what is OCR? Optical Character Recognition is the conversion of images of typed, handwritten or printed text into machine-encoded text. In a nutshell, this means taking an image of text, and turning it into text on a computer screen that we can edit.
A classic example of where OCR is used today is online checkouts that can “read” credit card information. Typically the checkout invites you to scan your debit or credit card card using your smartphone camera when completing a purchase. Then, rather than typing in the number, name and expiration date, the checkout will extract the key information and enter it straight into the form for you.
The checkouts using OCR have automated a dull, repetitive, manual task and reduced “fat finger” errors, which should improve checkout conversion. So how could this work in the mortgage journey?
When thinking about a mortgage application, there are a large number of fields that need to be completed by the advisor, most of which will need to be supported by some form of evidence. Not only is it a tedious process transferring information from document to computer screen, but as with all manual key entry tasks like entering a credit card number, errors are bound to crop up.
By using OCR in an application portal, CRM or sourcing system, brokers could quickly import details from images of documents without any typing required.
As a broker, imagine uploading a passport to a mortgage application portal where the name, date of birth and other key details are extracted and input into the form. Current addresses could be taken from utility bills and income information from a P60.
From the lender’s perspective, you also know that the details on the form have been cross referenced with the evidence provided. This could in turn reduce issues like duplicated credit searches stemming from typos and errors in applications.
A key theme in the industry over the last few years has been the movement towards reducing rekeying for intermediaries. One way of doing this is through APIs and integrating with other software providers in the journey. However, it can be expensive and time consuming to configure and build these connections, plus once they’re built you might also only have a solution for one other system in the journey.
With the use of OCR, the need to share information across systems is reduced. For example, if you as a lender were to have OCR configured to extract information from a fact find, brokers could upload a PDF fact find from any CRM system, and quickly autocomplete your application form from it.
Putting this all together, OCR could be used as a way of reducing rekeying without the need for integrating an API and reducing errors. Not only does this make an application significantly easier to complete for the broker, but should reduce time spent double checking and cross referencing the application and supporting documents, whilst incentivising brokers to attach relevant information upfront.
