We conducted an experiment to quantify the accuracy of methods based on n-alkanes and long-chain fatty alcohols for determining the diet composition of animals grazing complex swards. We cut forage from two indigenous vegetation communities, a Molinia caerulea-dominated grassland and a Calluna vulgaris-dominated dwarf-shrub community, and offered it to mature ewes in different ratios in a zero-grazing experiment. Nine dietary categories were identified within the forage offered: Molinia caerulea, Festuca spp., Juncus effusus, Carex spp., Calluna vulgaris, Erica tetralix, Vaccinium myrtillus, and dead grass. Samples of each of these categories together with fecal samples from each individual animal were analyzed for n-alkane and long-chain fatty alcohol concentrations. We analyzed the data using optimization software to minimize the sum of squares differences in the proportional profiles of n-alkanes and fatty alcohols in the diet and feces. Different combinations of n-alkane and fatty alcohols were investigated to assess which gave the most accurate measures of diet composition from the fecal profile. The most accurate estimates were obtained using combinations of the n-alkanes C25, C29, C31, and C33 and the long-chain fatty alcohols 1-C24-ol, 1-C28-ol, and 1-C30-ol, and these gave values for Lin's concordance correlation coefficient between estimated and actual values of >0.98. Our results demonstrate that n-alkanes and long-chain fatty alcohols can be used to estimate several components within the diet of animals grazing complex swards. Diet composition information obtained using this methodology has wide-ranging applications in terms of the assessment of the impact of grazing animals on particular ecosystems or the quantification of nutrient supply to the animal from different selection choices.