We explore when and how to reward failure in a dynamic principal-agent relationship with experimentation. The agent receives flow rents from experimentation, and divides his time between searching for evidence of success and failure about the underlying project. The principal commits in advance to rewards conditional on the type of evidence. At each instant, the principal makes a firing decision. We show that the principal’s optimal equilibrium features a stark reward structure—either the principal does not reward failure at all or rewards success and failure equally.