The economic feasibility of the use of rapid milk progesterone assays (RMPA) for early detection of pregnancy status in cows, to reduce the cost associated with reproductive inefficiency, was evaluated by use of decision analysis techniques. A decision tree was formulated and used to compare the RMPA to the alternative of palpation per rectum alone for the determination of pregnancy status. A computer spreadsheet was used to examine the effect of manipulating the values of inputs affecting the optimal decision (ie, cost of a day a cow is not pregnant, conception rate, cost of the RMPA, test sensitivity and specificity, cost of insemination) over the range of values likely to be encountered. The sensitivity of the optimal decision was greatest to the variables for cost of a day a cow is not pregnant, conception rate, and cost of the RMPA. When the cost of a day a cow is not pregnant is +2.00 and the cost of the RMPA is +6.00, the best decision is to use the RMPA when the herd conception rate is between 20 and 77%. When the herd conception rate is 60%, the best decision is to use the RMPA as long as the cost of a day a cow is not pregnant is greater than +1.44. Indifference curves were calculated, allowing both the cost of a day a cow is not pregnant and conception rate to vary, and used to select the most economical decision. Use of the RMPA for early detection of pregnancy status is likely to be economically advantageous in most dairy operations.