Low nutrients have been reported in potato-growing areas of Kenya, prompting a need for nutrient management research. A study was designed to determine the effect of omitting nutrients on potato growth, yield and harvest index. On-farm nutrient omission trials were set during the long rains (LR) and short rains (SR) of 2016 in which the treatments involve the judicious omission of N, P, K, S and B. Additional two treatments were included with one receiving all the nutrients and a control where no nutrients were added. The treatment was laid in a randomized complete block design with three replications. Potato yields reduced by 6.6 and 11.2 t ha−1 in N-omitted treatments in LR and SR, respectively, when compared to the one receiving all the nutrients, while omitting P resulted in respective yield reductions of 3.8 and 2.0 t ha−1. Stability analysis revealed that omission of N was more stable with a regression coefficient of 0.5; it was followed by P with a value of 1. Potassium, S and B were limiting nutrients only in some farms. N and P should continue to be included in potato nutrient management, while K, S and B should be added based on soil test.