The eastern Sahel region is plagued by low food security owing to climatic factors and continued social instabilities. This study aimed to forecast primary productivity (approximated here by the Normalized Difference Vegetation Index (NDVI)) by using antecedent primary productivity, rainfall, evapotranspiration, temperature, soil water amount, clay content and land use/land cover (LULC) data as predictors. A Random Forest model was used to forecast primary productivity one to six months ahead. The results showed correlations between observed and predicted primary productivity exceeding 0.91 for all months and forecast times. The forecasts showed antecedent primary productivity to be the most important predictor, followed by evapotranspiration and rainfall. LULC contributed moderately to the predictions with most LULC types exhibiting their changing importance with time. The study showed the ability of antecedent vegetation and climatic data, as well as the Random Forest algorithm, to predict primary productivity up to six months ahead. Such capability can inform the preparedness of communities at broad spatial scales in the Sahel region, which continues to suffer from climate variations and social instabilities.