The objective of the work was to identify local feeding strategies in small-scale dairy production systems during the rainy season in the highlands of Mexico, and to determine their effects on milk yields (MY), milk composition and economic viability. Twenty-two dairy farms were monitored by monthly visits, recording and sampling milk from between two and six cows in each farm, live-weight was also recorded. Samples from feeds used in that month were taken and feeds given to the dairy herd were weighed. Economic data was also recorded. Milk composition and milk urea nitrogen were determined, as well as chemical composition of feeds. Eighteen feedstuffs were identified, grouped in: HNH feeds-high in neutral detergent fibre (NDF) and in DM matter; HNL feeds-high in NDF but low in DM; HCh feeds-high in non-fibrous carbohydrates; and HCP feeds-high in crude protein. Four feeding strategies were identified: strategy 1 uses HND, HNL and HCP; strategy 2-HND, HNL, HCh and HCP; strategy 3 HNH and HCP; and strategy 4 HNL and HCP. Of participating farms, 73.4% followed strategy 1, 11.3% strategy 2, 11.3% strategy 4 and 3.8% strategy 3. There were no statistical differences (P > 0.05) between strategies for MY and milk composition, but there were differences (P < 0.05) for ration costs. Multiple regression analysis showed no significative (P > 0.05) model relating intake of feed groups and milk fat content, but milk protein and SNF contents were significantly explained by intake of HCP. When expressed as MY and milk components yield, milk fat yield was significantly explained by intake of all four feed groups, but milk protein and SNF yields were explained only by intake of HCP and LW. MUN excretion was explained also by HCP intake. All feeding strategies produced positive economic returns, on average generating the equivalent of 3.45 minimum wages of the area.