A wealth of climate forecast information and related prediction products are available, but impediments to adoption of these products by ranchers and farmers in the Unites States remain to be addressed. Impediments for agricultural applications include modest forecast skill, limited climate predictability, inappropriate forecast scale for site-specific applications, difficulties in interpretation of probabilistic forecasts by farmers and integration into agricultural decision systems, uncertainty about the value and effect of forecast information in multi-variable decision system, and generally low frequency of relevant forecasts. Various research institutions have conducted case studies of climate effects on agricultural production systems, particularly effects of historical ENSO signals in the south-eastern United States. Several studies addressed risk and economic values of seasonal climate forecasts, and others bridged the gap between current forecasting software and products and agricultural applications. These studies attest to the availability and suitability of forecast and impact-prediction software, as well as derived products for agricultural applications. Yet, little attention has been given to operational and application-specific prediction products for general agricultural use, and to an effective and affordable delivery system that reaches and resonates with the agricultural end-user (a prerequisite for adoption). The two latter impediments are the focus of this paper. Two existing approaches, the top-down and the participatory end-to-end approach for development and delivery of prediction products, are reviewed. A third approach, the hybrid approach, is emphasised and uses the top-down approach for climate forecast delivery and a participatory approach for development and delivery of farm-specific prediction information for the agricultural end-user. Suitability of such prediction products for agricultural applications and constraints to successful adoption are also discussed.