During the loan process: Reluctant to abandon manual review

Business Management

The loan stage is viewed as the risk taker and operator after credit approval, serving as the link between pre-loan and post-loan risk transmission.

◎ Risk Control Model Establishment

From the feedback results, all 16 surveyed consumer finance institutions mentioned building real-time credit approval systems through technologies such as artificial intelligence, cloud computing, and big data, while three institutions use a combination of traditional manual processes and risk control systems.

Dynamic Risk Management

Self-built Digital Infrastructure

◎ Debt Repayment is the Focus of Risk Control

Based on the information provided by the 16 consumer finance institutions, in the loan stage, they comprehensively assess users’ repayment ability based on multiple dimensions such as historical credit, asset status, and consumption stability.

Comprehensive Assessment of Repayment Ability

Multidimensional Data

Constructing balanced access and pricing-related complex risk models and strategy systems in the loan stage relies on advanced machine learning algorithms and rich data.

◎ Data Usage and Collection

In terms of data sources, the 16 surveyed financial institutions mainly adopt a deep integration approach of internally accumulated massive user data and foreign exchange market data. Leveraging the borrower data accumulation advantage, they perform deep data mining based on complex business scenarios and vast data (603138) to gather various risk data of customers.

Precise Refinement of User Profiles

Multi-source Data Collection

◎ R&D Progress and Achievements

According to feedback from the 16 institutions, due to differences in scale and revenue, there are also significant disparities in R&D investment and technological achievements.

Significant Anti-fraud Results

Polarized Patent Achievements

Business Development Challenges

In addition to differences in technological investment, each consumer finance institution has different insights into the difficulties faced in loan operation and their solutions.

◎ Evaluation Data Is Not Yet Complete

Currently, domestic income, debt, and credit data are still incomplete. Consumer finance institutions lack effective data support when assessing users’ repayment capacity.

Solution: Continuously introduce accurate third-party income or debt data, develop income and debt verification models, and achieve rapid and effective verification of borrowers’ repayment ability.

◎ Contradictions Between “Universal” and “Preferential”

Against the backdrop of overall interest rate reductions in the current consumer finance industry, the contradictions between “universal” and “preferential” finance are becoming more apparent. Increasing market competition also demands higher standards for refined management of existing customers, including more precise pre-emptive risk interception and control, and enhancing user stickiness.

Solution: Continue to promote digitalization, improve customer acquisition efficiency through technological means, reduce manual costs, and address operational difficulties with technology.