We characterized responses to an enzyme treatment commenced during the pre-calving period using pens in 3 large dairy herds. It was hypothesized that variation in intake of dietary feed components (g/d) sampled weekly would influence milk yield, fat and protein responses of milking pens; that there would be differences in responses between pens fed rations treated with a mix of fibrolytic enzymes or not and that dairy of origin would influence responses. Time-series cross-correlation analysis by pen was used with those results being pooled by meta-analytical methods to produce effect size (ES) estimates. The study provides unique insights to responses to enzyme treatment with feed and milk production responses being assessed over at least 3, and often, 6 mo. Differences in response (P < 0.05) between treatment and control pens occurred for soluble protein (ES = 0.249) on the same wk, ADICP (ES = 0.293) and lignin (ES = 0.237) both 1 wk before with milk protein %, and for ADICP (ES = 0.276) and lignin (ES = 0.246) 1 wk before with milk protein yield. These differences are consistent with treatment improving the digestibility of feed, particularly the protein and fiber fractions. Differences in production responses to intake of feed components among dairies were most frequently observed for the protein and fiber associated components. There were many more significant differences among dairies, than for between treatments. Differences between dairies were only observed once between the Californian dairies (Dairies 2 and 3), but >20 times for the South Dakota dairy (Dairy 1). The ES of differences between dairies were larger, on an absolute basis, than those between treatments and explained more than 7% of the variance (ES >0.5 or < −0.5) in production outcomes, in some cases. The greater ES likely reflect a greater difference in response to dietary component intakes among dairies than the difference in response achieved by enzyme intervention. The dairy that responded most to treatment had an estimated metabolizable protein (MP) excess resulting from diet, while the least responsive dairy had an estimated MP deficit and was the highest producing.