Daily milk production and reproductive performance of cows vaccinated with a live double-deleted Bovine Viral Diarrhoea Virus (BVDV) vaccine were compared to those of non-vaccinated cows, cohabitating in endemic BVDV herds. All animals in the treatment group were vaccinated on study day 0 irrespective of lactation or gestation status, while control animals did not receive any treatment. 1463 animals were enrolled in the study from four different farms in three different countries (UK, Italy, France). Endemic presence of BVDV in study herds was demonstrated by the detection of BVDV in the bulk tank milk, and seroconversion was evaluated at the beginning of the study. For individual animals, the day of calving was taken to be the start of lactation for the calculation of days in milk (DIM). The standard lactation period of 305 days was divided into three periods: early lactation (EL, from DIM 8 to DIM 102), mid lactation (ML, from DIM 103 to DIM 204 and late lactation (LL, from DIM 205 to DIM 305). For each farm and each lactation period, a mixed model statistical analysis was performed with daily milk production as response, and group, day as well as the interaction between those two factors as fixed factors. Chi-square test was used to compare abortion rate and prolonged inter-oestrous interval rate between treatment and control groups. A significant increase in milk production in the vaccinated group was observed in farms 1 (1.023 L/day) and 3 (0.611 L/day) during EL (p<0.001) and in farm 2 (1.799 L/day) during ML (P<0.001). In addition, at farm 2, vaccinated cows produced more milk than non-vaccinated cows starting from 80 DIM. No differences were found between groups in abortion rates or prolonged inter-oestrous interval rates. Data demonstrate that cows in herds endemically infected with BVDV and vaccinated with live double-deleted BVDV vaccine produce more milk; the difference in milk production occurs during early lactation.