The present study was aimed to compare automated heat watch system (AHWS) with visual observation (VO) for detection of estrus in Holstein Friesian cows. In this study, total 76 estrus and 1520 non-estrus days of 38 cows were used over period of 42 d. The frequency of VO method was 6 times in a day with 3 h interval. The duration of observation, at each time, was an hour. Heat was counted in VO on the basis of estrus behavior scoring. Within AHWS (Nedap livestock management, Groenlo, the Netherlands), cows had either smart leg (SL = 12) or neck tags (NT = 21). Cows underwent trans-rectal ultrasonography thrice weekly for ovarian status and during estrus every 3 h, to estimate the time of ovulation. A true estrus period was defined as 1) presence of large follicle, absence of corpus luteum on ovary 2) uterine tone. When detected heat, from either method, was coincided with a true estrus period defined as true positive (TP) otherwise false positive (FP). When no estrus was detected on true non-estrus days declared as true negative (TN) otherwise false negative (FN). Sensitivity was calculated as TP/ (TP + FN) *100%. PPV was calculated as TP/ (TP + FP) *100% and specificity was calculated as TN/ (TN + FP) *100%. Estrus detection efficiency was defined as number of true estrus/ ((total cows in breeding herd x days in period)/21) * 100. Data for efficiency of estrus detection, sensitivity, specificity, positive and negative predictive value for both methods, was analyzed by using Chi-squared (SPSS, IBM version) and compared. Results revealed that the efficiency of estrus detection was significantly higher (P < 0.05) in VO (x4) than in VO (x2), (73% (56/76) vs. 50% (38/76) respectively. Furthermore, the efficiency of estrus detection in NT (93%, 36/39) or SL (89%, 21/24) was significant (P < 0.05) higher than VO (≤x3; 63%, 48/76). The sensitivity and negative predictive value was significantly higher (P < 0.05) in AHWS than VO. It is concluded that detection of estrus by AHWS is better than VO. This is promising because of labor cost involved in VO method.