Recent changes in the National Polar-orbiting Operational Environmental Satellite System (NPOESS) program have levied a broad range of stringent requirements on automated cloud mask analyses to be derived from data collected by the Visible/Infrared Imagery/Radiometer Suite (VIIRS). These requirements are based on a review of user needs for NPOESS cloud products along with an assessment of the capabilities to retrieve automated cloud data products in an operational environment. As a result, four performance indicators are now used to evaluate analyses created with the VIIRS Cloud Mask (VCM) algorithm. They include a probability of correct typing (PCT), leakage rate, defined as the probability a pixel is classified as cloud-free when a cloud is present, false alarm rate defined as the probability a pixel is classified as cloudy when no cloud is present, and the capability to discriminate between clouds and heavy aerosols. Extensive testing of the VCM algorithm has now been completed at Northrop Grumman Space Technology and the predicted performance of the VCM algorithm has been determined. The result of this testing is the topic of this paper. The predicted performance is based upon analyses of globally distributed datasets of MODIS granules, compared to manually-generated cloud analyses, as well as testing with global synthetic VIIRS datasets. The results show that the wide range of NPOESS and user requirements can be satisfied with the VCM algorithm. In addition, a discussion is presented of activities that remain as NPOESS team prepares for the first VIIRS launch on NASA's NPOESS Preparatory Project (NPP) mission.