Abstract Nutrient enrichment with phosphorus (P) and nitrogen (N) threatens biodiversity globally, particularly because it drives invasion by exotic plant species. However, the effects of nutrients on plants can interact with other ecosystem processes such as competition and grazing, and these interactions can be scale‐dependent. Furthermore, P and N are often correlated, making it difficult to separate the effects of each. After implementing a herbivore‐exclusion experiment for 3 years in grassy woodlands, SE Australia, we assessed the relationship of native and exotic plant species with soil nutrients, grazing and spatial scale. Nitrate (NO3), available P and organic carbon (C) were partitioned into two principle components and further divided into plot‐ and site‐level variation. The correlated components of NO3, P and C (principle component 1) appeared to drive a dominant competitive relationship. Two exotic annual grasses likely out‐competed native and exotic species in areas high in NO3, P and C. In contrast, responses to principle component 2 (gradient from high NO3 to high P) indicated that native and exotic species generally benefitted when NO3 was high relative to P. Plant–nutrient relationships were further modified by grazing and spatial scale. Native and exotic species benefitted from grazing but did so under specific nutrient regimes. Summed occurrence of exotic species and native richness was reduced in the absence of grazing and high values of NO3 relative to P at site scales, but not at plot scales, suggesting competition is more important at a scale of hundreds of square metres while soil properties are most important at the plot scale (25 m2). Synthesis and applications. Dividing soil nutrients into uncorrelated principal components showed that niche space for native and exotic plant species is partitioned by positive and negative covariation of nitrate (NO3), phosphorus (P) and carbon (C), and interactions of these soil properties with grazing and spatial scale. Increasing NO3 relative to P could increase species richness, but requires further experimentation to define beneficial concentrations. Native plant richness can be enhanced with adequate levels of grazing by native herbivores and, in the long term, by reducing P from areas enriched with NO3 and P. These actions will help prevent likely competitive dominance by annual exotic grasses.