One of the difficulties groups have when making quantitative judgments is determining the relative quality of members’ inputs. Outcome feedback was investigated as a method for improving the ability of groups to identify their most accurate member. Results indicate that groups given outcome feedback were better at identifying their best member than were groups that practiced without feedback. This occurred even though the same individual was seldom the most accurate across items. However, groups given feedback were not as accurate as their best member any more often. This suggests that groups given feedback are learning how to discriminate valid from invalid cues, but are not using this information consistently when making their group estimates.