Existing data-driven methods for benchmarking building energy consumption often suffer from limited accuracy when evaluating the effectiveness of energy-saving measures. To address this challenge, we propose a novel approach based on Distance Metric-based Learning (DML). By leveraging distance metrics to identify similar days, the proposed method enhances the precision of energy consumption benchmarking. Experimental results show that our approach significantly outperforms traditional methods in evaluation accuracy. This work contributes to the advancement of reliable data-driven techniques for energy benchmarking and offers a robust tool for assessing and optimizing energy-saving strategies.