Controlled feeding experiments with juvenile abalone Haliotis discus hannai were performed to evaluate the suitability of stable isotope analyses for diet inference. Juvenile abalone were divided into five size groups with distinct isotopic ranges caused by different dietary regimes before the experiments. Three different foods were provided for 42 days: a benthic diatom Cylindrotheca closterium (BD), a macroalga Laminaria japonica (MA), and a formulated pellet (FP). For the FP diet, the smallest juvenile group [8.3 mm in mean initial shell length (SL)] showed significant changes in carbon and nitrogen stable isotope ratios from the natural diet within approximately 1 wk. Using final tissue-diet isotopic differences in the two largest juvenile groups for the FP diet, the fractionation values were estimated as 1.6‰ and 2.2‰ for δ¹³C and δ¹⁵N, respectively. An exponential decay function of the change in δ¹³C with time indicated that the shortest period to complete equilibrium fractionation was >120 days. Given the estimated fractionation values, the major diets of juvenile abalone (1.9–6.9 mm in SL) in the field were inferred to be benthic microalgae and several red macroalgae.