<p>The widespread adoption of smart meters has significantly made available detailed energy consumption data, providing valuable insights into consumer usage patterns. This paper takes advantage of this fact and presents a comprehensive time series analysis of energy usage patterns using smart meters time series data from high-usage power consumption companies in Tanzania. To the best of our knowledge, the study builds upon no any other data-driven research that uniquely focuses on TANESCO smart meter data to draw insights and forecast usage trends. Therefore, a systematic and scientific study is initiated to explore energy usage trends to enhance the power services by understanding customer consumption behavior. Our study introduces robust methods to handle missing data, a prevalent issue in Tanzania smart meter datasets due to power interruptions, ensuring robust energy consumption analysis. It, however, addresses the challenges of inconsistent data availability and significant discrepancies in power usage among different companies in Tanzania by developing methods to work with fragmented data and draw meaningful conclusions about long-term energy trends. The study devises an approach to unify discontinuous datasets for time series studies and analyzes variations in energy consumption patterns to reveal operational and sectoral influences on energy demand, providing insights into the underlying factors driving these variations and their implications for energy management. Our findings suggest that a detailed understanding of temporal consumption patterns can lead to more efficient demand-side management, encompassing improved demand response, power management, power monitoring, and distribution efficiency, resulting in cost-savings and a possibility of enhanced integration of renewable energy sources.</p>