top of page

AI Case Study

Ford Credit improves customer loan risk predictions over traditional credit risk scoring using machine learning

Ford Motor Credit Company compared its loan assessment model to ZestFinance's, which incorporates machine learning, and found the Zest model was better at predicting customer risk for automobile financing. Subsequently it plans to introduce machine learning for assessing creditworthiness in production.

Industry

Financial Services

Banking

Project Overview

"Ford Motor Credit Company and ZestFinance today announced the results of a study that measured the effectiveness of machine learning to better predict risk in auto financing and potentially expand auto financing for millennials and other Americans with limited credit histories. The machine learning study compared results from a Ford Credit scoring model with a machine learning model developed by ZestFinance using its underwriting platform to do deeper analysis of applicant data."

Reported Results

"The study showed improved predictive power, which holds promise for more approvals, enhanced customer experiences and even stronger business performance, including lower credit losses... machine learning-based underwriting could reduce future credit losses significantly and potentially improve approval rates for qualified consumers, while maintaining its consistent underwriting standards. As a result of the study’s success, Ford Credit is developing plans to implement machine learning credit approval models to further enhance its consistent and prudent lending practices across the credit spectrum."

Technology

Ford's proprietary scoring model and ZestFinance's ZAML; "ZestFinance is now offering the Zest Automated Machine Learning (ZAML™) Platform, which it developed specifically for credit underwriting. ZAML uses complex algorithms to analyze thousands of data points to provide a richer, more accurate understanding of all potential borrowers, delivered in an easy-to-use web interface. The ZAML Platform consists of three components: data collection and assimilation, machine learning modeling tools, and transparency tools that enable companies to explain credit decisions."

Function

Operations

General Operations

Background

"According to the U.S. Consumer Financial Protection Bureau, 26 million American adults, or about one in 10, have no credit record, making them difficult and often impossible to underwrite using traditional methods. This includes millions of millennials who are also part of the fastest-growing segment of new car buyers. Although these consumers may have steady jobs, their creditworthiness is heavily based on credit history. This makes it more difficult for companies to provide financing, and they could miss an opportunity for revenue growth. Ford Credit’s proprietary models have performed well for decades and the company is an industry leader in automotive risk management. Machine learning-based underwriting will be a game-changer for lenders, opening entirely new revenue streams. Millennials offer the perfect example. They are typically a good credit risk and are expected to command $1.4 trillion in spending by 2020, but many lack the financial history needed to pass a traditional credit check.”

Benefits

Data

Loan scoring model data

bottom of page