Technical Abstract: The majority of the world's ecosystems are managed to provide ecosystem functions and services on which we rely. Even with recent advances, ecosystem managers lack a clear methodology for applying basic knowledge of plant traits to meet management goals. Questions: Can we develop a methodology to address plant trait-function relationships in species-level units, which are missing from measures of community functional diversity but needed for management? Location: Northeastern U.S. grazing lands. Method: We measured 53 physiological, morphological, and growth traits for 19 plant species from greenhouse and field experiments. We developed a two-stage iterative method to link species to ecosystem processes related to management goals of increasing annual aboveground net primary productivity (ANPP) and its seasonal distribution in Northeastern U.S. grazing lands. Results: Species were distributed continuously, rather than clustering into discrete functional types. Grasses, legumes, and forbs overlapped considerably in trait values with these common classifications failing to adequately distinguish functional differences. Factor analyses were used to assess variation in species traits, and to rate species for six processes through which species contribute to annual ANPP. Species performed well in some processes and poorly in others, illustrating complex tradeoffs. Conclusions: This methodology provides a foundation for developing tools to guide construction of communities for applied settings, and for assembling hypotheses about plant functioning in mixtures.