The method of randomization has been a major driver in the recent rise to prominence of empirical development economics. It has helped uncover patterns and facts that had earlier escaped attention. But it has also given rise to heated debate and controversy. This paper evaluates the method of randomization and concludes that, while the method of randomization is the gold standard for description, it is not able to demonstrate causality. Nor does it, in itself, lead to policy conclusions, as is often claimed by its advocates. To get to policy conclusions requires combining the findings of randomized experiments with human intuition, which, being founded in evolution, has innate strengths. Moreover, even non-randomized empirical methods combined with reasoned intuition can help in crafting development policy.