A fairer and more personalised credit decision

Traditionally, banks make their credit decisions based on demographic signifiers and questionnaires to understand an applicant’s income and loan service capabilities. However, these demographic signifiers and questionnaires do not necessary reflect the applicants’ actual financial behaviour. The traditional credit scoring system also lacks of transparency.

Open banking has now made it possible to model actual financial behaviour based on transactional banking data. To do this, a new credit decisioning model is needed to meet unique needs of loan applicants and overhaul the lending system of the French financial industry.

To achieve these goals, Algoan used Kubernetes Engine to build a unique credit scoring platform that allows them to

- access raw banking transaction data via open banking, 

- crunch through high volumes of banking transaction data,

- develop new and more sophisticated ML models for credit scoring.

This new platform allows bank assess tailored loan application decisions based on customers’ actual banking transaction data for individuals. Compared to the traditional method of calculating credit scores based on questionnaires, new credit ML models are more powerful and personalised for loan applicants.

Outcome 

A fairer and more personalised credit decision making that is reflected on the complex financial needs in 21st century.