An experiment was conducted during 6 weeks to evaluate effects of a reduced dietary level of protein-rich concentrates in a moderate dairy production system on cows’ performance, protein efficiency and milk quality including fatty acid profiles. Twenty-three lactating cows (Swiss Fleckvieh) were assigned either to a group receiving on average 2·4 kg/d individually fed concentrates (Prot+, n = 12) or to a group receiving no individually fed concentrates (Prot−, n = 11). All cows had ad-libitum access to a total mixed ration (TMR) mainly based on grass and maize silage, hay and little potatoes and soybean cake. In weeks 4–6 of the experiment, part of the hay was excluded from the TMR, and fed separately in the morning. Individual feed intake and milk yield were recorded during weeks 3 and 6 of the experiment; at the same time feed, faeces and milk samples were collected twice per week for analyses. Data were processed in linear mixed models. Omission of individual concentrates in Prot− was fully compensated by higher roughage intake in terms of dry matter. Crude protein (CP) and net energy intake was almost maintained. Despite a lower apparent CP digestibility in Prot−, the ratio of milk protein to ingested CP was the same in both groups, indicating a higher ruminal utilisation of degraded CP in Prot−. This corresponded with lower milk urea concentrations in Prot−. Milk quality was affected in terms of lower concentrations of linoleic and conjugated linoleic acid in milk fat of Prot−. Concentrations of odd- and branched-chain fatty acids in milk were increased in Prot−. Sequential offer of hay and TMR did not lead to considerable effects in intake, efficiency and milk quality. In conclusion, the results indicate that the efficiency of feed protein utilisation for milk protein is not impaired if concentrates are reduced in a moderate- to low-input dairy production system.