Although typically modeled as a centralized firm decision, pricing often involves multiple organizational teams that have decision rights over specific pricing inputs. We study team input decisions using comprehensive data from a large U.S. airline. We document that pricing at a sophisticated firm is subject to miscoordination across teams, uses persistently biased forecasts, and does not account for cross-price elasticities. With structural demand estimates derived from sales and search data, we find that addressing one team’s biases in isolation has little impact on market outcomes. We show that teams do not optimally account for biases introduced by other teams. We estimate that corrected and coordinated inputs would lead to a significant reallocation of capacity. Leisure consumers would benefit from lower fares, and business customers would face significantly higher fares. Dead-weight loss would increase in the markets studied. Finally, we discuss likely mechanisms for the observed pricing biases.