Pesticide exposure is an important driver of bee declines. Laboratory toxicity tests provide baseline information on the potential effects of pesticides on bees, but current risk assessment schemes rely on one species, the highly social honey bee, Apis mellifera, and there is uncertainty regarding the extent to which this species is a suitable surrogate for other pollinators. For this reason, Osmia cornuta and Osmia bicornis have been proposed as model solitary bee species in the EU risk assessment scheme. The use of solitary bees in risk assessment requires the development of new methodologies adjusted to the biology of these species. For example, oral dosing methods used with honey bees cannot be readily applied to solitary bees due to differences in feeding behaviour and social interactions. In this study, we describe the “petal method”, a laboratory feeding method, and validate its use in acute and chronic exposure oral tests with Osmia spp. We conducted five experiments in which we compared the performance of several artificial flowers combining visual and olfactory cues against the petal method, or in which variations of the petal method were confronted. We then use the results of these experiments to optimize the feeding arenas and propose standardized methods for both acute and chronic exposure tests. The petal method provides high levels of feeding success, thus reducing the number of bees needed. It works with a wide variety of petal species and with both female and male Osmia spp., thus ensuring reproducibility across studies. To validate the use of the petal method in ecotoxicology tests, we assess the toxicity of a standard reference insecticide, dimethoate, in O. cornuta adults and determine LD50 values for this species. The petal method should facilitate the inclusion of solitary bees in risk assessment schemes therefore increasing the protection coverage of pesticide regulation.