United Trust Bank is using facial recognition technology to speed up its mortgage application process. The bank, which launched the service this week through a partnership with technology firm Nivo Solutions, is trying to eliminate the customer pain point of verifying identity, a requirement for mortgages in Britain.
United Trust claims that identity verification is a costly burden for customers, who must turn to third-party lawyers that charge fees and are only open during business hours. By relying on facial recognition, the bank hopes to create an end-to-end digital solution where customers can complete the entire mortgage application without having to leave the comfort of their homes.
According to Buster Tolfree, commercial director of mortgages at United Trust, the bank aims to increase its mortgage originations through an easier customer experience. “If you’re looking at mortgage providers, there’s often not a lot of difference in interest rates or payments, so one of the things that does matter is the friction within the journey,” he said. “[Now], they can do the journey in 90 seconds, and they can do it at midnight on a Sunday night.”

According to United Trust, customers would spend anywhere from £100 to £350 ($125 to $439) in fees to verify their identity through lawyers before implementing the facial recognition feature. Customers would need to schedule a meeting with a lawyer and commute to the lawyer’s office whenever they had to apply for a mortgage. Tolfree said verifying identity with a lawyer used to take about 30 minutes, not including travel.
Using facial recognition, customers can verify their identity directly in United Trust’s app without having to see any third parties, and the verification process takes about 90 seconds. Once customers have chosen a United Trust mortgage, a chatbot in the app will prompt them to take a picture of their government-issued ID. Customers then are prompted to take a short selfie video where they must read a series of letters or numbers and move their face around, and the app ensures the face in the selfie video matches the ID picture. Once customers have verified their identity, they can continue filling out the rest of the application through the app or online. In addition to speeding up the customer journey, Tolfree said the tool cuts down on fraud that comes from forged documents and unlicensed lawyers attempting to take advantage of people.
See also: As New Technologies Falter, Identity Remains a Top Concern for Businesses
United Trust began looking at facial recognition after hearing from brokers that the verification process was fraught with difficulty for customers. The bank piloted the facial recognition feature over the past few months, but Tolfree didn’t disclose how many people were involved in the pilot. The bank officially rolled out the feature to customers on Wednesday.
Although Tolfree admitted it’s still early days for facial recognition-based digital identity verification, he said more than 90% of customers so far are able to finish the process in 90 seconds or less. United Trust doesn’t market its products to end users, instead relying on its partnerships with licensed brokers who promote the bank’s products. The tool is available to anyone applying for their first mortgage, as well as certain second mortgage customers applying for more than £100,000 (about $125,000).
While facial recognition technology has the capability to reduce mortgage application times, it also presents some risks. Mo Chaudhury, an associate professor of finance at McGill University, said facial recognition raises some privacy concerns. “If banks using facial recognition share or sell such data without the permission of the applicant, it will aggravate the already serious problem of privacy in the online world of doing business,” he added.
Bank Innovation Build, on Nov. 6-7 in Atlanta, helps attendees understand how to “do” innovation better. It is designed to offer best practices, to guide the innovation professional to better results. Register here and save with early bird rates ending September 27th.






