To optimize treatments for gastrointestinal nematode (GIN) in adult dairy cows by selective treatment, we need to identify cows with an improved milk production (MP) after treatment. The objective of this study was to evaluate the evolution of MP after treatment for GIN, and to identify herd and individual variation factors. A field trial involving 20 pastured dairy herds in Western France was conducted in autumn 2010 and autumn 2011. In each herd, lactating cows were randomly allocated to a treatment group (fenbendazole) (533 cows), or a control group (544 cows). Daily cow MP was recorded from 2 weeks before until 14 weeks after treatment. Individual serum anti-Ostertagia antibody levels (expressed as ODR), pepsinogen levels, egg output, and bulk tank milk ODR were measured at time of treatment. Moreover, in each herd, information regarding heifers’ grazing history was collected to assess the time of effective contact in months (TEC) with GIN larvae before the first calving. MP data were analyzed using linear mixed models (herd and cow as random effects). The overall treatment effect on MP was significant but slight (maximum=+0.86 kg/d on week 6 after treatment). Cows from herds where the percentage of positive egg count was >22.6% responded better than those from herds where it was lower. Cows from low-TEC (<8 months) versus high-TEC herds also showed a better response, particularly when the bulk tank milk ODR was high; however, bulk tank milk ODR taken into account alone did not appear as a significant variation factor. Primiparous cows, cows with days in milk (DIM)<200 at time of treatment, and cows with low individual ODR (<0.38) responded better than multiparous cows, cows with DIM>200, and cows with higher ODR respectively. These results highlight promising key criteria for selective treatment for GIN in dairy cows.