Benchmarking uses information of similar organizations or processes to assess the gap between current and achievable performance. In the context of agriculture, benchmarking usually refers to a systematic comparison of the performance of farms producing under similar bio-geophysical conditions.
Benchmarking assesses performance in relative, rather than in absolute terms. This is particularly suited for assessing effects of agricultural management, because the strong influence of external factors such as climate, weather or soils on farm performance is difficult to isolate. Simple benchmarking identifies best practices from a set of samples and then evaluates all samples relative to these best practice examples. In more complex benchmarking where multiple evaluation criteria are applied, samples are not evaluated relative to a single best practice but relative to a so called efficiency frontier, which represents the best performances considered achievable and which is calculated based on data from all samples. The most common parametric and non-parametric approaches to this are Stochastic Frontier Analysis (SFA) and Data Envelopment Analysis (DEA) respectively.