Mastitis is a major challenge for the dairy industry. The current study aimed to evaluate the performance of Draminski^® mastitis detector (DMD) and ultrasonography as cow-side tools for detecting bovine subclinical mastitis. To achieve this purpose, a total of 103 raw milk samples, including 48 samples from buffaloes and 55 from cows were tested using DMD, the California Mastitis Test (CMT), ultrasonography, and bacterial culture. Based on the results of CMT and bacterial culture, the study defined the quarter health status (QHS) as healthy, latent, or inflamed. The study evaluated the performance of the DMD and ultrasonography via Spearman’s rank correlation and calculated the cut-point value for DMD, in addition to identifying the sensitivity, specificity, and predictive values for ultrasonography against QHS. Results showed that percentages of quarters defined as healthy or inflamed using QHS categorization were higher in buffaloes than cows, unlike the latent ones. The study found a significant inverse correlation between DMD and QHS (– 0.695 in buffalo and – 0.526 in cow samples). Using inflamed/non-inflamed QHS classification, the cut-point values for the DMD were 515 for buffaloes and 465 for cows. The diagnostic performance analysis of ultrasonography against inflamed/non-inflamed QHS revealed a sensitivity of 83% and specificity of 22%. The negative predictive and positive predictive values for ultrasonography against inflamed/non-inflamed QHS were 78% and 29%, respectively, and the overall accuracy was 39% as a screening method for subclinical infection. The study concluded that DMD is suitable for early detection of bovine subclinical mastitis. The cut-point value assigned by the DMD manufacturer is potentially low. Therefore, the study suggests a cut-point value of 515 for buffaloes and 465 for cows. Finally, ultrasonography is more appropriate for identifying structural intramammary changes rather than for screening subclinical infections‎‎‎.