The amount of supplementary food humans provide to wild animals is increasing, yet the full effects of this provisioning remain unclear. For these effects to be assessed at the levels of both the population and the individual, the degree to which individuals are using this resource must be quantified. Traditional approaches rely on observations of feeding animals and analysis of gut contents and feces, but these have several limitations. Stable-isotope analysis can overcome some of these. If supplementary food items are isotopically distinct from natural ones, the relative contribution of supplementary food to the diet may be quantified accurately. We demonstrate how the isotopic signature of supplementary foods can be manipulated to increase their discrimination from natural food sources and provide an example of the utility of this approach in a supplementary feeding study. We provided supplementary food over a winter, then sampled birds during the following breeding season and analyzed their claws for their isotopic signature to estimate diet choices. The results highlight considerable variation in individuals' use of supplementary food, both within a study site and between different sites. Often the results from supplementation experiments are inconclusive. Even within the same species there can be an effect in one year or location but not in others, so a method for quantifying variation in food uptake could help in interpretation of the results. Stable-isotope analysis allows the effects of experimentally increased food supplies on ecology and behavior to be assessed accurately.