The deployment of Remote Monitoring System (RMS) based on Internet of Things (IoT) technologies and data management in the cloud is a cutting-edge solution for water management agencies to track harmful algae blooms (HABs). However, each environment presents different intrinsic characteristics, hence the reliability and representativeness of collected data may vary according to the selected RMS architecture. This study aims to implement the same RMS in two different water bodies in Spain: the freshwater As Conchas reservoir and the shallow L’Albufera brackish water lagoon. We firstly evaluate the pros and cons of current technologies available on the market and then we select an RMS based on a combination of plug-and-play (YSI-based) and customisable solutions (Libelium-based). The deployment of nodes on buoys allows big data collection from different areas of water bodies, their continuous transfer and storage on a web server, and their subsequent visualisation on the web interface in real time. Secondly, in order to know whether the selected RMS architecture and monitored environment influence the representativeness of the collected data, we perform a Pearson correlation test between the deployed nodes and satellite images. The results point out that the more heterogeneous the environment is, the more nodes must be deployed in different areas for a longer time to obtain a realistic view of the water body status. Therefore, this study provides critical and empirical data to implement a profitable and effective real-time monitoring system in other HAB-affected areas.