We compared sources of variability and cost in paired stomach content and stable isotope samples from three salmonid species collected in September 2001–2005 and describe the relative information provided by each method in terms of measuring diet overlap and food web study design. Based on diet analyses, diet overlap among brown trout, rainbow trout, and mountain whitefish was high, and we observed little variation in diets among years. In contrast, for sample sizes n ≥ 25, 95% confidence interval (CI) around mean δ15Ν and δ13C for the three target species did not overlap, and species, year, and fish size effects were significantly different, implying that these species likely consumed similar prey but in different proportions. Stable isotope processing costs were US$12 per sample, while stomach content analysis costs averaged US$25.49 ± $2.91 (95% CI) and ranged from US$1.50 for an empty stomach to US$291.50 for a sample with 2330 items. Precision in both δ15Ν and δ13C and mean diet overlap values based on stomach contents increased considerably up to a sample size of n = 10 and plateaued around n = 25, with little further increase in precision.