Documentation of the energy yield of a large photovoltaic (PV) system over a substantial period can be useful to measure a performance guarantee, as an assessment of the health of the system, for verification of a performance model to then be applied to a new system, or for a variety of other purposes. Although the measurement of this performance metric might appear to be straightforward, there are a number of subtleties associated with variations in weather and imperfect data collection that complicate the determination and data analysis. A performance assessment is most valuable when it is completed with a very low uncertainty and when the subtleties are systematically addressed, yet currently no standard exists to guide this process.This report summarizes a draft methodology for an Energy Performance Evaluation Method, the philosophy behind the draft method, and the lessons that were learned by implementing the method. The general philosophy behind the methodology includes the following features:• The method is performance-model agnostic.• The performance model must not be inadvertently modified, when being implemented on the measured meteorological data sets, relative to the model that was used on the historical data set.