Business and financial records for 2001 from 228 New York dairy farm businesses are summarized and analyzed. This analysis demonstrates the use of cash accounting with accrual adjustments to measure farm profitability, financial performance, and costs of producing milk. Traditional methods of analyzing dairy farm businesses are combined with evaluation techniques that show the relationship between good management performance and financial success. The farms in the project averaged 277 cows per farm and 21,762 pounds of milk sold per cow, which represent above average size and management level for New York dairy farms. Net farm income excluding appreciation, which is the return to the operator's labor, management, capital, and other unpaid family labor, averaged $149,044 per farm. The rate of return including appreciation to all capital invested in the farm business averaged 12.2 percent. Differences in profitability between farms continue to widen. The top 10 percent of farms average net farm income excluding appreciation was $693,355, while the lowest 10 percent was a negative $16,775. Rates of return on equity with appreciation ranged from 43 percent to negative 14 percent for the highest 10 percent to the lowest 10 percent of farms, respectively. Farms adopting bovine somatotropin (bST) attained higher rates of milk production per cow, had larger herds and in all but one year were more profitable than farms not adopting bST. Farms adopting rotational grazing generally produced less milk per cow than non-grazing farms but had somewhat lower costs of production and higher profitability. Large freestall farms averaged the highest milk output per cow and per worker, the lowest total cost of production and investment per cow, and the greatest returns to labor, management and capital. Farms milking three times a day (3X) were larger, produced more milk per cow and had higher net incomes than herds milking two times per day (2X). Operating cost per cwt. of milk was $0.08/cwt. higher for 3X than 2X milking herds, while output per cow was 4,342 pounds higher. One should not conclude that adoption of these technologies alone were responsible for differences in performance.