When designing a bench test to model wear in lubricated contacts associated with engine and gearbox applications, considerations of scale are important. It is hazardous simply to attempt to define the “real life” conditions (load, speed, temperature, etc.) and glibly apply them to test pieces on a bench test machine. If large amounts of energy are being dissipated in small test specimens with supporting structures that do not allow the heat to escape, then it is clear that the specimens will become very hot. The bulk temperature may exceed what is experienced in practice, producing transitions in wear or frictional response. This paper outlines the relationship of wear and failure to energy dissipation in terms of two global parameters, Matveesky's Friction Power Intensity and Plint's Energy Pulse, and applies these to previously published data. It then shows how these can be used in the design of appropriately scaled, simple bench test procedures for the evaluation of lubricants for two practical EHD lubricated contacts (gears and cam/follower).