Credit risk

Leverage innovation to bring newfound speed and accuracy to credit risk modeling

The results of the global survey reveal great uncertainty in the modeling of credit risk,
highlighting the need for AI, machine learning and alternative data

By Kim Miner, Senior Vice President, Global Marketing at Arise from

With all the disruption caused by COVID-19 over the past two years, how robust are credit risk models? This was one of the questions we sought to answer with a global research study that interviewed 400 industry decision makers. The results were more than a little troubling: Only 18% of fintech companies and financial services organizations believe their credit risk models are accurate at least 75% of the time.

This is amazing – especially considering that the rest of the respondents – in fact the vast majority of respondents – indicated that they believed their credit risk models to be accurate. less more than 75% of the time. Especially since credit risk modeling is at the heart of every fintech and financial services company.

This state of high uncertainty in credit risk modeling exposes the shortcomings of legacy approaches to credit risk decision-making that leverage limited data, workflow, and automation – often in separate systems. To truly improve decision-making, organizations need more data, more automation, more sophisticated processes, more forward-looking predictions, and faster decision-making. And to do that, they need AI, machine learning, and alternative data.

While artificial intelligence, machine learning and alternative data may have been on the credit risk decision list a few years ago, fintechs and financial services organizations are quickly realizing that the he alternative – legacy technology and approaches – simply don’t measure up to today. credit risk decision task.

Our survey highlighted the growing appetite for AI predictive analytics and machine learning, data integration and the use of alternative data as a way to improve credit risk decision making. . Real-time credit risk decision-making was the top area of ​​investment respondents planned for in 2022, as the organization works to address the current “financial fault line” in credit risk decision-making. credit risk.

Financial services executives see AI-powered risk decision-making as the cornerstone of improvements in many areas, including fraud prevention (78%), automating decisions across the credit life cycle (58%), improved cost savings and efficiency (57%), more competitive pricing (51%) and improved accuracy of credit risk profiles (47%) .

However, many companies are struggling to mount the resources needed to support their AI initiatives; AI development and implementation can be time-consuming, and cost-prohibitive. Only 21% of financial services organizations begin to see ROI from AI initiatives within 120 days.

Leveraging innovation to bring newfound speed and accuracy to credit risk modeling 7

Fraud continues to grow for financial services and lending companies, both before and during the pandemic, with identity fraud losses reaching $56 billion in 2020. And while great progress has been made in When it comes to financial inclusion, there is still a long way to go with an estimated 1.7 billion adults remaining globally unbanked.

Sixty-five percent of decision makers in our survey indicated that they recognize the importance of alternative data in credit risk analysis for better fraud detection. Additionally, 51% acknowledge its importance in supporting financial inclusion