valU, the Leading Buy-Now, Pay-Later (BNPL) Lifestyle Enabling Fintech Platform, Launches its Rules Engine 2.0 Featuring Enhanced Risk-Assessment Algorithms for Higher Approval Rates
The new automated tool features an enhanced credit risk assessment-based algorithm that will allow for higher approval rates through streamlined machine decision-making.
valU, the MENA’s leading and multi award-winning Buy-Now, Pay-Later (BNPL) lifestyle enabling fintech platform, announced today the launch of its new Rules Engine 2.0 for Higher Approval Rates. The new automated tool features an enhanced credit assessment-based algorithm that will allow for higher approval rates through streamlined machine decision-making.
“The launch of our enhanced Rules Engine 2.0 reflects our ongoing commitment to expanding our unrivalled risk-assessment methodologies, ensuring that valU’s service offering remains aligned with the fast-paced fintech industry and maintaining our position as a market leader,” said Mostafa El Sahn, Chief Risk Officer of valU. “As one of the MENA’s leading BNPL lifestyle enabling fintech platforms, continuously developing and improving our credit scoring model is crucial to ensuring we maintain the highest levels of risk management in line with global standards without compromising on the customer experience as we work to meet the ever-growing needs of our clients. The innovative automated credit assessment system will allow valU to broaden its acceptance rate, increasing it from 55% to 80% due to its machine decision-making. This innovation also features greater emphasis on credit assessment, with a clear focus on performance and scalability, with additional functional enhancements,” concluded El Sahn.
The new rules engine is designed to rank customers by condensing a variety of variables and attributes into a single score. It will approve valU limits subject to each customer’s risk rating, guaranteeing a more accurate risk-based assessment. The new rules engine will also have higher precision by handling all outliers and missing values through grouping similar attributes with similar predictive strengths, increasing from the model’s accuracy in the process.