© Daly et al.; licensee BioMed Central. Background: Schmallenberg virus (SBV) is a recently emerged virus of ruminants in Europe. Enzyme-linked immunosorbent assays (ELISA) are commonly used to detect SBV-specific antibodies in bulk tank milk samples to monitor herd exposure to infection. However, it has previously been shown that a bulk tank milk sample can test positive even though the majority of cows within the herd are seronegative for SBV antibodies. Development of a pen-side test to detect antibodies in individual milk samples would potentially provide a cheaper test (for which samples are obtained non-invasively) than testing individual serum samples by ELISA. Therefore, the aim of this study was to investigate the agreement between antibody levels measured in milk and serum. Results: Corresponding milk and serum samples from 88 cows in two dairy herds in the UK were tested for presence of immunoglobulin G antibodies to SBV using a commercially-available indirect ELISA. A serum neutralisation test (NT) was also performed as a gold standard assay. The ELISA values obtained for the bulk tank milk samples corresponded with the mean values for individual milk samples from each herd (bulk tank milk values were 58% and 73% and mean individual milk values 50% and 63% for herds A and B, respectively). Of the 88 serum samples tested in the NT, 82 (93%) were positive. Although at higher antibody levels, the ELISA values tended to be higher for the individual milk samples than for the corresponding serum samples, the positive predictive value for milk samples was 98% and for serum samples 94%. The serum ELISA was more likely to give false positive results around the lower cut-off value of the assay. Conclusions: The results indicate that testing of individual milk samples for antibodies against SBV by ELISA could be used to inform decisions in the management of dairy herds such as which, if any, animals to vaccinate.