This article analyzes the demand for a simple rainfall-based weather insurance product among farmers in rural India. We explore the predictions of a standard expected utility theory framework on the nature of demand in terms of price, the basis of the hedge, and risk aversion using data from a randomized control trial. We find that demand behaves as predicted: it falls with price and basis risk and is hump-shaped in risk aversion, with price sensitivity decreasing at higher levels of basis risk. We estimate a negative price elasticity of 0.58 and find that doubling the distance to a reference weather station decreases demand by 18%. These results indicate that improving pricing and quality of insurance products can directly increase demand. In addition, we examine the impact of insurance training relative to other mechanisms designed to increase understanding. The evidence suggests that increased incentives to learn or learning by using are more effective at increasing both understanding and demand. Finally, we contribute to the scarce evidence on the demand for insurance over time. In terms of our main interventions, we find that the effect of premium subsidies persists over time, while the impact of investments in new weather stations diminishes and the effect of increased training in the first season seems to disappear during the second season. Importantly, while having previously purchased insurance does not encourage future uptake, receiving a payout does. This could reflect issues of trust in the product or the insurance company, and constitutes an important topic for future research.