Rising residential electricity demand, driven by population growth and appliance proliferation, strains non-renewable energy resources and increases carbon emissions. In India, households consume ~25% of total electricity, yet most users receive only monthly aggregated bills without real-time insights or control over individual loads. Existing commercial energy monitors are costly (>USD 100) and lack remote access, whereas open-source alternatives (Raspberry Pi, Arduino) suffer from complexity, high power use, or missing cloud integration. This work presents a low-cost (<USD 20), web-based, self-hosted energy management system using NodeMCU and PZEM-004T V3.0, enabling real-time monitoring, remote control, and data-driven optimization for sustainable residential energy use. This study presents an efficient energy management system for residential buildings, addressing the increasing demand for electrical energy and the need for improved management practices. The system integrates a NodeMCU (ESP8266) microcontroller and a PZEM-004TV3.0 module to monitor and manage electrical energy loads. Real-time data on voltage, current, power, and energy were acquired and transmitted to Google Firebase via the MQTT protocol, enabling remote monitoring and control. Users can manage appliances locally through manual switches and light-emitting diode indicators or remotely through a web interface. The system achieved a calibration efficiency of 98.58% compared to standard instruments, with voltage and current errors of 0.15% and 3.12%, respectively. By providing detailed insights into energy usage and control capabilities, this cost-effective solution empowers users to optimize consumption, reduce wastage, and contribute to energy sustainability efforts.