Understanding and controlling complicated experimental systems well enough to achieve reproducibility is hard. Lining up incentives to encourage scientists to tackle this challenge together, rather than abandoning it as unhelpful in the career competition, might make reproducibility easier.