In response to increasing integration of renewable energy sources on electric grid systems, battery energy storage systems (BESSs) are being deployed world-wide to provide grid services, including fast frequency regulation. Without mitigating technologies, such as BESSs, highly variable renewables can cause operational and reliability problems on isolated grids. Prior to the deployment of a BESS, an electric utility company will typically perform modeling to estimate cost benefits and determine grid impacts. While there may be a comparison of grid operations before and after BESS installation, passive monitoring typically does not provide information needed to tune the BESS such that the desired services are maintained, while also minimizing the cycling of the BESS. This paper presents the results of testing from a live grid using a method that systematically characterizes the performance of a BESS. The method is sensitive enough to discern how changes in tuning parameters effect both grid service and the cycling of the BESS. This paper discusses the application of this methodology to a 1 MW BESS regulating the entire island of Hawaii (180 MW peak load) in-situ. Significant mitigation of renewable volatility was demonstrated while minimizing BESS cycling.