The recent focus on impact evaluation within development economics has lead to increased pressure on aid agencies to provide "hard evidence", i.e. results from randomized controlled trials (RCTs), to motivate how they spend their money. In this paper I argue that even though RCTs can help us...