Working with Heron Financial, Data Scientists from the the Hartree Centre North East Hub created an AI-powered mortgage decision engine to help mortgage advisors save time on manual qualification tasks to allow them more time with clients.

‍

Challenge

Heron Financial, one of the UK’s most productive mortgage brokerages, faced a critical operational challenge constraining growth and client conversion. Despite their market-leading performance and proprietary CRM platform, mortgage advisers were spending significant time on manual client qualification tasks, including affordability assessments, credit data interpretation, and suitability matching. Potential clients who were ‘not ready yet’ often fell through operational cracks due to inconsistent triage. With approval rates varying significantly across advisers (65-85%) and geographic regions, the company recognised the need for data-driven standardisation. 

‍

Approach

Recognising the potential of AI and machine learning to transform their mortgage advisory process, Heron Financial partnered with Data Scientists from the Hartree Centre North East Hub to develop a predictive model that calculates early-stage qualification insights. The team integrated and processed historical data from Heron’s Everglades CRM system, utilising 54,678 anonymised historic cases with their outcomes from 2019-2025. Through data preprocessing and feature engineering,  Data Scientists created over 40 predictive variables encompassing assets, financial profiles, property intelligence, and credit risk indicators. Multiple machine learning algorithms were evaluated, with XGBoost selected for its superior performance. SHAP and KNN clustering was then implemented to identify positive and negative factors influencing each prediction, enabling advisers to understand qualification factors while ensuring regulatory compliance and user trust.  

‍

‍

Benefits

By automating early-stage mortgage assessments, this solution delivers a reduction in manual qualification time. Beyond operational efficiency, the AI revealed critical business insights, including credit over-utilisation, lender relationship patterns and visa conditions. Importantly, the tool acts as an intelligent assistant to Heron’s mortgage advisers, allowing them to focus on client relationships and complex case handling rather than routine data processing. With plans to integrate the system into a chat assistant for automated early-stage client interactions, this enables Heron to scale operations, handle growing lead volumes from employee benefit partnerships, and ensure consistent, data-driven assessments while maintaining regulatory compliance through explainable AI decisions.

‍

“Working with the Hartree Centre North East Hub has been transformative for the business. This solution saves advisers valuable time on manual tasks and allows the team to focus on what they do best: building relationships with clients and providing advice.”

“The explainable AI approach gives us confidence in the decisions while maintaining the personal touch that defines service. We’re excited to be at the forefront of fintech innovation in the mortgage advisory sector, and we see this as a significant step towards scaling our operations while maintaining our industry-leading conversion rates.”

- Matt Coulson, Heron Financial

‍

Further Information

This work was completed as part of one of our collaborative data projects. The projects are up to 12 weeks in duration and give you access to a wide range of expertise across our team of data scientists and data engineers. We will work alongside your team to scope your data science or engineering project, build a prototype solution, and explore options to deploy it within your organisation. You can learn more about them on our webpage here.

If you would like to learn more about the Hartree Centre North East Hub or our collaborative data projects, please get in touch with us at: hello@hartreenortheast.uk

‍