This contribution is based on the notion that different technologies are present in an industry. These different technologies result in differential “drivers” of economic performance depending on the kind of technology used by the individual firm. In a first step different technologies are empirically distinguished. Subsequently, the associated production patterns are approximated and the respective change over time is estimated. A latent class modelling approach is used to distinguish different technologies for a representative sample of E.U. dairy producers as an industry exhibiting significant structural changes and differences in production systems in the past decades. The production technology is modelled and evaluated by using the flexible functional form of a transformation function and measures of first- and second-order elasticities. We find that overall (average) measures do not well reflect individual firms’ production patterns if the technology of an industry is heterogeneous. If there is more than one type of production frontier embodied in the data, it should be recognized that different firms may exhibit very different output or input intensities and changes associated with different production systems. In particular, in the context of localized technical change, firms with different technologies can be expected to show different technical change patterns, both in terms of overall magnitudes and associated relative output and input mix changes. Assuming a homogenous technology would result in inefficient policy recommendations leading to suboptimal industry outcomes.