Out of necessity, emission and fuel consumption test cycles are a simplified representation of the real-life use of a vehicle or component that is assumed to be most common. In reality, variations are introduced by both the driver and the environment - and to a lesser degree also by the vehicle itself through performance deviations because of tolerances in the components' characteristics. However, since such simplified test cycles exist and are accepted (or even required by law), OEMs tend to use them also in product development to benchmark their products against the competition, and to make decisions on how to optimize design.While this approach might give acceptable results for on-road vehicles, it fails to capture reality in the case of versatile working machines. Here, the variety of possible applications cannot be covered by one common application but rather demands a mix of several cycles. This has a large impact on the setup and evaluation of physical and virtual testing of working machines, especially those with alternative power systems like hybrids, which will be discussed in the paper.Attempts to simplify the complexity of real-life applications are made in some testing standards and standard proposals, but they over-simplify with the result of prescribing a common cycle that is not representative and which therefore, however tempting, cannot be used for any meaningful benchmarking of emissions and fuel consumption.