The value of diversity metrics to represent ecological communities and inform broad-scale conservation objectives and policy has often been subject to debate and uncertainty 1 and 2. In practice, diversity metrics are important in setting management and conservation priorities, just as economic indices contribute to global monetary and financial policies. Thus, key challenges for ecologists are to identify new ways to view and summarise patterns in biodiversity and improve on the metrics available for management purposes. In a recent paper on functional diversity patterns in reef fishes , we highlighted the potential of new insights gained from functional trait-based approaches to inform marine management, stressing the need to develop and refine biodiversity measures that are linked to ecology (rather than taxonomy). We used a unique, fisheries-independent reef fish identity and abundance dataset, collected using standardised methods from equatorial to high latitude regions all over the world, to provide the first global view of the distribution of individuals amongst species (including a measure of evenness) and functional traits amongst marine communities. A recent paper by Robinson et al.  published in Marine Policy criticised the use of our evenness index as a measure of biodiversity, and questioned the use of functional trait-based metrics derived from surveys of standardised areas for decisions relating to broad-scale management of marine systems. In this paper we respond to Robinson et al. and rebut their claims related to sampling bias and broad-scale applicability of trait-based approaches.