Using APIs to make the mortgage application process faster

lender Mar 31, 2021

Applying for a mortgage is complicated. Customers have brokers to help them when they’re looking to buy a property or remortgage an existing one – they navigate the minefield for you.

But when it comes to actually filling in that application, brokers are really just filling in a glorified paper form. We hear from lenders whose online forms haven’t changed in years, or cost thousands of pounds to update, or – even worse – aren’t online at all.

Mortgage applications are data-heavy, relying not only on data from the client, but also from other sources. Lenders will check the land registry to make sure the title deed is kosher. They’ll use credit checking services to ensure their client is creditworthy and doesn’t have problem debt. And the global pandemic has accelerated the use of AVMs (automated valuation models) to estimate the value of a property.

But all of these checks are done after the mortgage application is received, or at different steps that vary between each lender. The client uploads their documents, the broker hits submit, and then the lender has to gather all that data to make their lending decision. What if we could simplify some of this work, and gather that data as part of the application process itself?

We think that credit checks, AVMs, open banking and more can be built directly into the mortgage application form, giving brokers feedback on their application in real-time – and in some cases even making a decision on the application before it’s submitted.

Take property licensing. If you’re looking to buy a property to rent out and it has more than a certain number of bedrooms, you’re looking at a house of multiple occupancy or HMO. All HMOs above a certain size require a licence from the local council; this means additional requirements on the landlord and extra red tape.

Some lenders are reluctant to lend on these properties. That’s a problem if a broker hasn’t done their research – they might submit an application that’s not going to be accepted. One lender we spoke to rejects fully half of their buy-to-let applications on property licensing grounds.

What if they didn’t have to? We’ve integrated with a data-driven regtech company that provides us with detailed information on a property’s licensing requirements via their RESTful API. The broker types in the address, clicks a button, and we can tell them immediately if a licence is needed – and, with our Policy Engine, whether that’s a dealbreaker for the lender.

That means we can prevent that case from being submitted in the first place. The lender never even sees it – the broker knows it won’t be accepted, so is able to take their business elsewhere, happy that their time hasn’t been wasted chasing underwriters and BDMs for a credit decision, and the customer is saved a wasted application fee.

The lender’s happy, too – they didn’t have to process a case only to reject it outright. And with all those cases filtered out at the source, they’ve unlocked more capacity they can use to improve their service, lower their rates, or just issue more loans.

Pulling in data from an external API and integrating it into the application process is going to become commonplace. APIs are being offered by more companies: HMRC’s can be used to get an individual’s PAYE income, Xero’s to gather business earnings income. We can even use APIs to trigger actions – such as initiating a wire transfer on a client’s behalf to pay the fees on their new mortgage.

Our vision is for underwriters to receive a perfectly-packaged application, complete with all the data they need to make a credit decision. We’re collaborating with some leading fintechs to consume their APIs directly from our application form, and working to build these integrations around an open standard for fintech data services.

If you’re a UK lender and looking to cut time to offer, unlock operational capacity and offer a class-leading digital experience, get in touch!

Photo by Lars Kienle on Unsplash

Henry Stanley

Co-founder and CTO at Mast. Likes Ruby, Rails, Golang, 12-factor apps and TDD. Has taken the Founder's Pledge – a commitment to donate a portion of personal proceeds to high-impact charities.