BNP Paribas Leverages AI to Expedite Mortgage Approvals
BNP Paribas is utilizing artificial intelligence to streamline its mortgage application process, significantly reducing processing times and enhancing customer satisfaction. The new system, implemented in November 2024, uses Intelligent Document Processing (IDP) to automate key steps. This innovation aims to address the traditional challenges associated with mortgage financing, where gathering and verifying documents can be time-consuming and prone to errors.
Addressing the Challenges of Mortgage Applications
Sébastien Perrigault, Head of Retail Financing at Commercial & Personal Banking in France, highlighted the initial hurdles. “Obtaining a mortgage is one of the first steps. It can be a source of stress and delay, as well as dissatisfaction for our customers, who expect us to deal with their requests quickly.”
Previously, advisors manually reviewed extensive documentation, including tax returns, pay slips, and bank statements. This process, alongside manual data entry, led to longer processing times. The bank’s goal was to reduce the time required to issue a digital offer.
The AI-Powered Solution: IDP and Doc Factory
To overcome these obstacles, BNP Paribas integrated an AI solution to streamline the mortgage approval process, shortening approval times. This initiative incorporates IDP, specifically for basic mortgage loans, excluding more complex applications like bridging loans or real estate partnerships. This AI system checks the consistency of information between the documents and the data entered in the advisor’s credit planning tool, which reduces errors.
The AI solution is designed to read documents, extract relevant information, and verify their accuracy. For instance, it cross-references details from purchase agreements with borrower income data to ensure consistency. The AI also validates the authenticity of tax notices.
“We use classic AI, which we have combined with an internally developed IDP (Intelligent Data Processing) software called Doc Factory,” Perrigault explained. The IDP solution leverages Natural Language Processing (NLP) technologies, which are capable of processing large volumes of text from application documents and customer data. The AI has been trained on thousands of documents to identify and verify the data.
Benefits for Advisors and Customers
For advisors, the IDP acts as a “super assistant,” automating key steps in the file processing. The AI automatically reads over 80% of document data, with 60% automatically matched and validated. This results in a time savings of approximately 30 minutes per file. Moreover, the IDP helps to reduce risks by assuring document consistency and authenticity.
The implementation of AI has significantly shortened the traditional processes. “On the end-to-end ‘real estate’ path, the IDP therefore makes it possbile to save 5 days and grant a mortgage in less than 30 days.”
BNP Paribas aims to further improve the customer experience by issuing digital mortgage offers directly through branches for the simplest cases (roughly 20% of applications). This will cut down the time between the customer’s decision and the offer’s issuance by an additional three days.
The time saved will enable advisors to advise customers on financial solutions and provide customer support, while offering advice on energy renovation solutions
Optimized Customer Experience and Future Plans
The integration of AI improves data reliability and reduces errors, contributing to increased customer satisfaction. As Perrigault emphasized, “The ‘real estate’ path is highly valued by our customers and these technological developments allow us to further improve their satisfaction by offering an increasingly simple and fast experience.”
The project, which began in France in November 2024, currently supports 80% of customer real estate files. BNP Paribas is now developing new AI models to focus on additional documents, such as energy performance certificates and savings account statements, starting in 2025. The bank plans to launch the next big step in early 2026, with the possibility of issuing around 20% of offers electronically, which would significantly reduce processing times and costs.
