The efficient operation of heating, ventilation, and air conditioning (HVAC) systems is critical for both energy performance and occupant comfort in buildings. This paper presents a conceptual framework for a data-driven monitoring system designed to support real-time diagnostics, modelling, fault detection, and performance optimization in HVAC applications. The proposed architecture integrates a network of physical sensors—such as temperature, humidity, flow rate, heat flux, and energy consumption sensors—connected to distributed edge-processing units (“hubs”) capable of local data acquisition and inter-device communication. In addition to environmental measurements, the system monitors the operational status of key components (e.g., pump activity, valve position, on/off cycles) to provide insight into system behaviour and detect anomalies. These hubs continuously log sensor data and transmit it to a central platform where time-series storage, visualization, and higher-level analysis take place. The framework emphasizes modularity, scalability, and the ability to support decision-making through data interpretation and predictive techniques. Potential use cases include fault detection, dynamic system modelling, and energy efficiency evaluation. The paper also discusses sensor selection criteria, data architecture, and possible future extensions toward adaptive control strategies.