Agriculture is inherently a risky business and farmers face a variety of production and market related risks. The prevalence of risk in agriculture is particularly acute among small scale farmers who populate the often highly risky environments found in developing countries. Agriculture is particularly sensitive to climate change; future projections indicate that farmers will face more challenging conditions characterised by high crop price volatility, high crop yield variability and concomitant income risk (IPCC, 2013).  It is well documented that households that are unable to insulate themselves from risk avoid them by opting for lower return, lower risk agriculture, and thus do not benefit from technological innovation that can enhance long term productivity. The net effect is that risk contributes to low agricultural productivity and poverty.  In Kenya, crop and livestock insurance has been proposed as a risk management instrument to enable farmers to cope with weather related and other production risks. However, agricultural insurance in Kenya is imperfect or non-existent and there is a dearth of empirical evidence on its potential demand. This paper empirically investigates farmers’ demand for crop insurance in Kenya. Using survey data, we focus on the role that the design of insurance contracts can have on encouraging effective demand. We argue that demand for insurance will be stronger and more sustainable if, among other factors, the design is informed by farmers’ preferences.