With the development of the ecosystem approach to fisheries, improving fishing selectivity has increasingly been put forward as an objective for management. The aim of this paper is to clarify the limits of fishing selectivity and its use in fisheries management. Fishing selectivity would be better apprehended if restricted to the catching process only, not to the utilisation that is made of the catch once onboard, which falls under catch utilisation. Confusion would be further limited if fishing selectivity is restricted to the fishing operation scale, while exploitation pattern, i.e. the distribution of fishing mortality at the population or community level, applies to larger scales. Fishing selectively is minimizing bycatch – catching primarily the fishing trip targets. Since the ecological consequences of maximising the target catch relative to bycatch remain unknown at integrated scales, fishing selectivity cannot be used as an objective in itself. However, its small scale, high manageability and good understanding make it a convenient instrument to reach management objectives at large scales. Selectivity can serve to manage what is extracted from the ecosystem and thus what can be used, and/or to manage what is left in the ecosystem and how fishing impacts it. Different factors affect fishing selectivity, catch utilisation and exploitation patterns, some of them are manageable and thus can be used to move towards these objectives. Management tools are diverse, but need to be integrated to meet large-scale objectives. The complexity of dealing with large scales incurs a need to develop the available knowledge on exploitation patterns and catch utilisation to be able to adequately manage and monitor progress toward selectivity-related objectives.