Internal Search in Fabory

This thesis focuses on the design, implementation, and validation of measures aimed at improving internal search performance and product data quality in the Fabory e-commerce platform. The proposed solution addresses key challenges of B2B e-commerce, including low search relevance, incomplete product information, and the lack of systematic data governance. The approach combines search configuration optimization, the use of artificial intelligence for automated data enrichment, and the introduction of a structured data governance framework. The solution was validated through a pilot phase using real production data and evaluated based on business and process metrics, particularly Zero Search Results (ZSR) and Add-to-Cart rate (ATC). The results demonstrate that a data-driven approach supported by AI leads to measurable improvements in search relevance, commercial performance, and significant reductions in manual effort. The thesis also establishes a scalable and sustainable foundation for further automation and long-term development of product data management.