evidoria

← Back to browse

Good practice Imported

Smart Nkunganire System — Rwanda's AI-Assisted Fertilizer Recommendation Tool for Smallholder Farmers

Rwanda · Kigali · See the Rwanda profile · See the Kigali profile

Top 66% 53/100 · Ask Evidence Copilot about this practice

Rwanda's Agriculture Board channels over 90% of the country's subsidized fertilizer through the Smart Nkunganire System, whose CGIAR-built machine-learning models reached 97% accuracy recommending crop and fertilizer choices in field validation for rice and potato.

Smart Nkunganire System — Rwanda's AI-Assisted Fertilizer Recommendation Tool for Smallholder Farmers

Details

Promoter
Rwanda Agriculture and Animal Resources Development Board (RAB), with CGIAR's Excellence in Agronomy Initiative and One Acre Fund
Period
2018–2021 platform pilot; ML fertilizer recommendations scaling 2023–2026
Keywords
agriculture, digital agriculture, food security, agricultural extension

Description

Rwanda's Agriculture and Animal Resources Development Board (RAB) runs the Smart Nkunganire System (SNS), a digital platform that over 1.5 million registered farmers use to order subsidized agricultural inputs. Since 2023, RAB has worked with CGIAR's Excellence in Agronomy Initiative, technology partner BK Techouse and scaling partner One Acre Fund to layer a machine-learning fertilizer-recommendation tool onto SNS, using the CGIAR AgWise toolkit and multi-season, multi-location fertilizer-response trial data to generate site-specific advice for six priority crops: cassava, potato, rice, wheat, maize and beans.
CGIAR reports the underlying crop-and-fertilizer prediction model reached 97% accuracy in its validation trials, with rice and potato recommendations fully validated as of 2024 and the remaining four crops still in validation. An initial rollout reached 10,000 pilot farmers supported by 200 farmer facilitators and 10 sector agronomists, with a 2025 scaling target of roughly 300,000 farmers (20% of SNS's registered user base) receiving ML-based recommendations. Over 90% of Rwanda's subsidized fertilizer already moves through the SNS platform, though that figure describes platform-wide input distribution rather than the ML-recommendation feature specifically.
CGIAR's own reporting is unusually candid about the tool's limits: farmers describe a cumbersome USSD interface (regularly requiring dozens of steps), heavy reliance on agro-dealers and cooperative leaders rather than direct self-service use, delayed subsidy payments that disrupt affordability, connectivity-dependent stock-tracking discrepancies, and no coverage yet for organic fertilizer, lime or micronutrient recommendations. CGIAR explicitly flags sustainability risk if government subsidy support lapses.

Read the full analysis: https://www.topafricanews.com/2024/05/04/the-rwanda-sns-innovation-expansion-tailored-fertilizer-recommendations-for-smart-and-sustainable-farming/

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful