This paper explores the challenges and potential benefits of implementing Model Predictive Control (MPC) in buildings, particularly its ability to enhance energy flexibility and reduce costs. While MPC adoption is growing commercially, it remains difficult to determine whether energy cost savings justify implementation costs, especially for medium to large commercial buildings where one-size-fits-all solutions are impractical. The study highlights the value of load shifting in response to dynamic electricity prices as an additional proposition to improve MPC’s attractiveness. Experimental results from an office building in Berkeley demonstrate the effectiveness of MPC in leveraging building thermal mass for energy flexibility, achieving up to 50% cost savings compared to the existing controller under dynamic pricing scenarios.