Retrieval-Augmented Generation (RAG) systems rely on effective document ‘chunking’, a process developers often struggle to optimize due to inadequate tools for overview and analysis. This study investigates developer perceptions of chunk visualization for identifying, debugging, and resolving issues within the RAG development process. Applying Information Visualization and Explainable AI (XAI) principles, a visualization tool was designed and evaluated using a Design Science Research (DSR) approach. A prototype was iteratively developed and assessed through qualitative interviews with two experienced RAG developers, followed by thematic analysis. Findings indicate current methods are perceived as inefficient. The visualization tool, particularly featuring a side-by-side document/chunk view, was received positively. Developers reported it significantly enhanced understanding, aided identification of chunking problems (e.g., size, context, overlap), and improved debugging efficiency compared to existing practices. In conclusion, chunk visualization is perceived by developers as a valuable approach to increase transparency, efficiency, and problem-solving effectiveness in RAG system development, offering insights for creating more robust applications.