Sweet potato [Ipomoea batatas (L.) Lam.] varieties can be distinguished by their flesh and skin colors (white, yellow, orange, and purple). Loss of crop diversity is a particular concern in centers of origin and secondary centers of diversity where varieties have evolved with high adaptability to the local environment as well as in developing countries where farmers rely on crop diversity for food security. Therefore, the present study was undertaken to assess on farm diversity and status of genetic erosion of sweet potato in Haramaya district, Ethiopia. The source of population for the study was sweet potato cultivating small holder farmers Haramaya district. Data were collected for similarity/ dissimilarity index, preference index, and genetic erosion. The result of preference index indicated the highest preference index (25.60) with preference rank 1st was recorded for BeelaBochol genotype as the least preference index (12.10) was recorded for BeelaBaleyso genotype. The highest similarity coefficient (0.80) was observed between Bachake and Maya Gudo followed by similarity coefficient value (0.75) between Maya Guddo and Khuro. Itwas found that the genetic integrity was found to be 88.9% as genetic erosion was 11.1%. The high genetic integrity implies little genetic erosion or loss of sweet potato varieties in the study area. The farmers’ cultivation and utilization of sweet potato varieties in the study area demonstrated that most of the farmers (90%) plant one (selected) variety for cultivation, and know less than 5 (< 5) varieties of sweet potato. About 144 (66%) of the respondents use intercropping system for sweet potato cultivation. The majority of respondents 212 (97%) observed loss of sweet potato varieties. In the study area, 180 (82%) of the respondent farmers suggested the use of their own seed source for planting. About 160 (73%) of the respondents suggested the cultivation of sweet potato as subsistence crop. The major constraints of sweet potato cultivation as suggested by the respondents were tiresome land preparation 80 (37%), lack of improved variety 56 (26%), land scarcity 50 (23%), and market demand 33 (15%). All respondents 219 (100%) suggested farmers intention to continue cultivation of sweet potato. It can be concluded from the result of the present study that farmers can identify and prefer sweet potato genotypes frequently using storage root yield and boiled quality.