Buy Now Pay Later (BNPL) services such as Shopee PayLater have transformed consumer transaction patterns by offering convenience and flexibility in shopping. However, systems like this also carry high risks if not accompanied by adequate software testing. This article presents a study aimed at evaluating the reliability of the Shopee PayLater system using the Black Box Testing approach with Boundary Value Analysis (BVA) and Equivalence Partitioning (EP) techniques, combined with user-based use case scenarios. This method allows for testing various input variations, especially at boundary values and data structure validity. Out of 14 test cases executed, 11 successfully detected input errors accurately, while the remaining 3 failed to handle certain conditions, such as character length and OTP format. The Defect Removal Efficiency (DRE) obtained was 21.43%, indicating that the effectiveness of testing can still be improved. Overall, this approach has proven to enhance testing coverage and detect relevant potential errors in digital financial systems.