Reduction of livestock losses from predators is a complex problem that requires the integration of lethal and nonlethal management tools. During 2005 and 2006, we tested shock collars for conditioning wild wolves (Canis lupus) in Wisconsin, USA, to avoid bait sites over an 80-day period. Treatment wolves (n = 10) visited shock zones less and spent less time at shock zones compared to control wolves (n = 4) during 40-day shock periods and 40-day post-shock periods. Treatment wolves remained away from shock zones for a greater number of days compared to control wolves. A smaller proportion of treatment pack members visited shock zones during shock and post-shock periods compared to control packs. Shock collars conditioned treatment wolves to avoid bait sites for >40 days and reduced visitation by other pack members. We also demonstrated the application of shock collars at the scale of livestock farms. Shock collars could serve as a useful nonlethal tool for managing livestock depredations, particularly in chronic problem areas and with endangered populations. © 2012 The Wildlife Society.