We discuss the application of an end-use load shape estimation technique to develop annual energy use intensities (EUIs) and hourly end-use load shapes (LSs) for commercial buildings in the Pacific Gas and Electric Company (PG&E) service territory. The results will update inputs for the commercial sector energy and peak demand forecasting models used by PG&E and the California Energy Commission (CEC). EUIs were estimated for 11 building types, up to 10 end uses, 3 fuel types, 2 building vintages, and up to 5 climate regions. The integrated methodology consists of two major parts. The first part is the reconciliation of initial end-use load-shape estimates with measured whole-building load data to produce intermediate EUIs and load shapes, using LBL's End-use Disaggregation Algorithm, EDA. EDA is a deterministic hourly algorithm that relies on the observed characteristics of the measured hourly whole-building electricity use and disaggregates it into major end-use components. The end-use EUIs developed through the EDA procedure represent a snap-shot of electricity use by building type and end-use for two regions of the PG&E service territory, for the year that disaggregation is performed. In the second part of the methodology, we adjust the EUIs for direct application to forecasting models based on factors such as climatic impacts on space-conditioning EUIs, fuel saturation effects, building and equipment vintage, and price impacts. The core data for the project are