Abstract Indtroduction Currently, Ethiopia is following an agricultural development-led industrialization strategy with a major goal of helping agriculture grow so that it can encourage growth in other sectors of the country’s economy. However, it is characterized by low productivity due to technical and socioeconomic factors. To improve this problem, integration of modern technologies with improved level of efficiency becomes more crucial. Therefore, this study tries to fill the gap by investigating efficiency variations and factors affecting technical efficiency of red pepper production in North Gondar zone Amhara regional state, Ethiopia. Methods By using multistage sampling, cross-sectional data were collected from 385 systematically selected households. Stochastic frontier Cobb–Douglas production was estimated. Result and conclusion The results of the analysis revealed that a mean technical efficiency of red pepper was 78.80% (ranging from 16 to 94.9%). This implies that red pepper producers can reduce current level of input application by 21.2% given the existing technology level. The estimated stochastic production frontier (SPF) model also indicates that land, seed, chemical, oxen and labor are significant determinants of red pepper production level. The estimated SPF model together with the inefficiency parameters shows that age, education status, land size, land fragmentation, extension service, credit access and market information were found to statistically and significantly affect the level of TE of red pepper farmers in the study area. Hence, emphasis should be given to improve the efficiency level of those less efficient farmers by adopting the practices of relatively efficient farmers in the area so that they can be able to operate at the frontier. Specifically the concerned body should provide adult and vocational education for the farmers and create opportunities for farmers with lower technical efficiency to have experience and best practices sharing with those that scored efficiency scores close to one.