Professional Design of a Soil Data Analytics Platform for Smallholder Farmers in Borno State

Improving agricultural productivity in post-conflict Borno State requires data-driven soil management guidance that is accessible and actionable for smallholder farmers with limited formal education. This study designed a soil data analytics platform for smallholder farmers in Maiduguri, Konduga, and Bama Local Government Areas, Borno State, North East Nigeria. A professional design methodology was applied, drawing on agronomic soil data from 480 farm plot samples collected by CARI researchers, structured consultations with 18 agricultural extension officers and 10 data science for agriculture specialists, and review of soil analytics platforms deployed in comparable low-resource contexts in East Africa and the Sahel. The platform designed specifies a soil data ingestion pipeline for GPS-tagged field test results, a crop suitability recommendation engine using rule-based and machine learning classifiers calibrated to Borno soil types, a fertiliser recommendation module aligned with IFDC prescription models, and a farmer-facing interface accessible via USSD for feature phone users. Offline data synchronisation protocols for areas with intermittent connectivity are detailed. Expert review by nine soil science and agricultural data science specialists confirmed the platform's agronomic accuracy. The study recommends deployment through Borno State ADP extension network, training of 200 extension officers as platform ambassadors, and integration with the ADP input supply voucher system to connect soil recommendations to subsidised fertiliser procurement.

Keywords: soil analytics, smallholder farmers, Borno State, crop recommendation, agricultural data platform