• Bayesian Structural Time Series
• causal inference
• counterfactual estimators
• data science
• energy consumption
• energy efficiency
• energy insights platform
• Interrupted Time Series
• sustainability

This thesis explores the application of causal inference methodologies to measure the impact of Eliq's energy management solutions on end-users' electricity consumption. With the increasing demand for electricity and the imperative to reduce greenhouse gas emissions, understanding the causal effects of energy management solutions becomes indispensable. Leveraging causal inference techniques, this research investigates whether Eliq's solutions lead to changes, particularly reductions, in energy consumption among customers. Through counterfactual estimators such as Bayesian Structural Time Series and Interrupted Time Series models, the study analyzes multi-dimensional data collected from various locations to estimate the causal impact. Results demonstrate a significant negative causal impact in the majority of cases, indicating a reduction in electricity consumption attributed to Eliq's insights platform. However, a smaller portion of locations exhibit a positive causal impact, suggesting an increase in consumption due to factors not always controllable by Eliq. Overall, this research contributes to advancing the understanding of energy management solutions' effectiveness and their role in promoting energy efficiency and sustainability.