Innovative methods for development and application of index-based financial instruments for climate risk management in Africa are presented. The approach involves statistical modeling of climate variability and climate impacts, with a focus on drought. It is often noted that seasonal rainfall totals do not represent the optimum approach to evaluating climate impacts on crop yield. Methods to overcome this have been explored. Analyses have been undertaken using satellite estimates of vegetation greenness, effectively providing an integration of daily rainfall and other climate aspects relevant to vegetation growth. Indeed, estimation of crop yields through remote sensing techniques has been a major effort since at least the dawn of the satellite era, and its application in index insurance is perhaps a natural development from those studies. A complimentary approach is to use a transformation of daily station rainfall data, such as by driving a simple water requirement satisfaction index (WRSI) crop model. Each approach has its advantages and limitations, and the technically optimum way of developing climate indices for index insurance in Africa will likely represent a combination of such approaches. The methods are being applied to Millennium Village locations, therefore providing results in a variety of agroecological and climatic zones in Africa. We report drought frequencies and the ability of the different indices to trigger payouts in the drought years. We also report how these results are being used in the Millennium Villages Project.