Videos produced by and for farmers fill a critical information gap by providing smallholder farmers with access to relevant, actionable information with which to make farm-related decisions.

Key success factors include: videos’ highly localized content; human mediation to reinforce messages; capacity building that strengthens service provision; and near real-time data collection and analysis that informs adaptations. Partnering with existing extension providers allows us to tap into trusted networks operating in remote rural communities, enabling rapid scale-up. The approach has become a platform for collaboration between public, private and civil society actors to disseminate high-quality, consistent and relevant content, share learning, and influence national programs at scale. More than a message delivery vehicle, our approach has organized timely exchange of locally relevant knowledge, and strengthened the social structures that meet to discuss them.

We have facilitated production of more than 6,000 localized videos in 50 languages and dialects, which have been screened by 17,000 frontline workers to reach 2.3 million farmers (77% women) across 10 countries, primarily India and Ethiopia. Over 50% of viewers have adopted practices (3 to 4 on average), representing 3.5 million total adoptions. Adoption of selected practices has raised farmers’ production by 22% on average and their incomes by 16%.

A recent two-year randomized control trial (RCT) conducted by the International Food Policy Research Institute (IFPRI) in Ethiopia shows that the video-enabled approach reaches 24% more farmers than the public extension system’s conventional approach and results in up to 44% higher uptake of promoted practices. Adoption of improved practices appeared to spill over across the target area, even among those who did not view videos. Overall, adoption rates throughout kebeles (village clusters) in which videos were shown was 35%; and effects were nearly twice as high for those who viewed the videos (61%). Extension agents who use video make a greater effort to visit farms and provide follow-up advice than those who do not.

A peer-reviewed RCT conducted by Innovations for Poverty Action and Jameel Poverty Action Lab in Bihar, India, demonstrated a 50% gain in adoption rates and 21% increase in paddy production compared to the Government of Bihar’s Rural Livelihoods Promotion Society’s traditional group-based approach.

Based on project costs incurred by Digital Green (equipment, staff, training, etc.) over one agricultural season, the IFPRI-led RCT calculated the cost of securing one additional adoption of a common agricultural practice to be $4. These findings are similar to the Microsoft Research RCT that assessed the effectiveness of Digital Green’s approach in 2009 in India, which calculated the cost per adoption to be $3.70, compared to a conventional train and visit-style extension approach, which had a cost of $38.18 per adoption. A 2015 cost benefit study in Odisha, India, which measured the cost-effectiveness of the video approach in terms of knowledge retention of disseminated nutrition-related messages, found that the unit cost was $2.47 per successful retention of a promoted practice.

An internal study (2017-18) on identification and promotion of the top 2-3 agricultural practices with highest returns on investment for farmers for widely grown commodities in two states in India shows that farmers who adopt the identified practices have significantly higher yields than those who do not. Specifically, it shows yield differences of: 24% for paddy (Bihar); 68% for wheat (Bihar); 64% for potato in Bihar and 74% for potato in Jharkhand; and 328.6% for pigeon pea (Jharkhand). Potato and wheat farmers in Bihar doubled their incomes from those crops, with potato farmers earning an additional $120 and wheat farmers earning an additional $108.