In this episode, Product Owner Faik Özcan shares the architectural journey behind Kuveyt Türk’s real-time fraud detection product, which analyzes nearly 2.5 million transactions daily. In financial institutions, the ability to analyze relationships between large-scale data, uncover historical connections, and perform these operations in real time is crucial for preventing fraud. While Relational Database Management Systems (RDBMS) offer ACID compliance, data integrity, and standardization, their rigid schema and vertical scalability limits can hinder agility in rapidly evolving environments.
Özcan explains how the team transitioned to a NoSQL document-oriented model using MongoDB, enabling flexible schema design, horizontal scalability, and high availability. The shift improved query performance, reduced response times by nearly 50%, and simplified integration with modern development practices such as agile and code-first approaches.
Although challenges such as unlearning relational design habits, adapting to JSON-based query syntax, and managing database operations emerged, the move resulted in faster development cycles, centralized yet flexible data modeling, and better system resilience. Özcan concludes that adopting NoSQL in fraud detection not only enhanced performance and scalability but also built a foundation for more adaptable, future-ready financial systems.
To review the article and access more blog posts 👉 Architecht – Blog