A study, sponsored by Provenir, a fintech providing AI-powered risk decisioning software, has outlined the greatest credit risk analysis challenges, opportunities, and trends fintech decision-makers see in 2022.
The study also shows the growing appetite for AI predictive analytics and machine learning, data integration, and use of alternative data as the means to improve credit risk decisioning and support the key imperatives of fraud detection/prevention and financial inclusion.
Provenir surveyed 400 decision-makers in fintech and financial services organizations across North America, Latin America, Asia Pacific, Europe and the Middle East and found only 18 per cent of fintechs and financial services organisations believe their credit risk models are accurate at least 75 per cent of the time.
“Consumer credit markets have changed dramatically over the past two years, yet many financial services organisations are still employing legacy approaches to credit risk decisions. The net result is that organisations today have a substantial level of uncertainty in the accuracy of their risk models which results in less inclusive credit, fewer approvals, and reduced opportunity for business growth,” said Larry Smith, CEO and Founder of Provenir.
This ‘risky business’ uncertainty in credit risk modelling accuracy may be why real-time credit risk decisioning was survey respondents’ No. 1 planned investment area in 2022. Additionally, the survey shows organisations are recognising the value of AI and machine learning, alternative data, and data integration in credit risk decisioning approaches.
AI-enabled risk decisioning is seen as key to usher in improvements in many areas, including fraud prevention (78 per cent), automating decisions across the credit lifecycle (58 per cent), improving cost savings and efficiency (57 per cent), more competitive pricing (51 per cent) and improving the accuracy of credit risk profiles (47 per cent).
The survey also gauged how organisations want to use alternative data in credit risk analysis; improving fraud detection and serving the underbanked/unbanked were the top main objectives cited. 65 per cent of decision-makers polled recognise the importance of alternative data in credit risk analysis for improved fraud detection. Additionally, 51 per cent recognise its importance in supporting financial inclusion, 43 per cent see its value in expanding target markets, and 40 per cent say its use results in more accurate credit scoring.
Despite strong recognition of the value of alternative data, many organisations struggle with operationalising alternative data within their credit risk models. Data integration was cited as the biggest impediment to the use of alternative data by seven out of 10 respondents.
According to the study, organisations are also looking to lean into the latest technological advancements in their automated credit risk decisioning platform selection:
- AI – 55 per cent of respondents who plan to invest in an automated credit risk decisioning system consider AI to be one of the most important features.
- Low-code/no code approach – 80 per cent of respondents consider a low/no code user interface critical.
- Model language interoperability – 42 per cent cited model language interoperability as key.
- Utilisation of alternative data sources – Nearly half (48.5 per cent) of those planning to invest in automated credit risk decisioning systems this year say improved utilisation of alternative data sources is an important feature.
