Vegetation response to nutrient addition can vary across space, yet studies that explicitly incorporate spatial pattern into experimental approaches are rare. To explore whether there are unique spatial scales (grains) at which grass response to nutrients and herbivory is best expressed, we imposed a large (∼3.75 ha) experiment in a South African coastal grassland ecosystem. In two of six 60 × 60 m grassland plots, we imposed a scaled sampling design in which fertilizer was added in replicated sub-plots (1 × 1 m, 2 × 2 m, and 4 × 4 m). The remaining plots either received no additions or were fertilized evenly across the entire area. Three of the six plots were fenced to exclude herbivory. We calculated empirical semivariograms for all plots one year following nutrient additions to determine whether the scale of grass response (biomass and nutrient concentrations) corresponded to the scale of the sub-plot additions and compared these results to reference plots (unfertilized or unscaled) and to plots with and without herbivory. We compared empirical semivariogram parameters to parameters from semivariograms derived from a set of simulated landscapes (neutral models). Empirical semivariograms showed spatial structure in plots that received multi-scaled nutrient additions, particularly at the 2 × 2 m grain. The level of biomass response was predicted by foliar P concentration and, to a lesser extent, N, with the treatment effect of herbivory having a minimal influence. Neutral models confirmed the length scale of the biomass response and indicated few differences due to herbivory. Overall, we conclude that interpretation of nutrient limitation in grasslands is dependent on the grain used to measure grass response and that herbivory had a secondary effect.