Dairy cows in modern production systems are at risk to develop metabolic disorders during the transition period. Reasons for individual differences in susceptibility, as well as the underlying pathomechanisms, are still only partially understood. The development of metaphylactic treatment protocols is needed. In this context, an on-farm prospective 3-fold blinded randomized study involving 80 German Holstein cows was performed throughout 1 yr. The trial involved a thorough recording of the production and clinical traits, clinical chemistry, and liver biopsies and blood and urine sampling at d 14 (mean: 12 d, range: 1-26 d) antepartum (AP), and d 7 (7, 4-13) and 28 (28, 23-34) postpartum (PP) for metabolomics analyses. Two groups received a treatment with butaphosphan and cyanocobalamin (BCC) at either the dosage recommended by the manufacturer or the double dosage (5 or 10 mL/100 kg of body weight 10% butaphosphan and 0.005% cyanocobalamin (Catosal, Bayer Animal Health), n = 20 in each group, parity: 4.2 ± 2.0 and 3.4 ± 1.3, respectively (mean ± SD)] and one group a placebo treatment (NaCl 0.9%, n = 40, parity: 4.0 ± 1.9). The animals were treated at 6 time points (7, 6, and 5 d AP, and 1, 2, and 3 d PP) via intravenous injection. Mass spectroscopy-based targeted metabolomics analysis of blood plasma and liver samples were performed using the AbsoluteIDQ p180 kit (Biocrates Life Sciences), whereas the urine samples were analyzed by nuclear magnetic resonance spectroscopy. Statistical analysis was performed using multivariate [partial least squares discriminant analysis (PLS-DA)] and univariate methods (linear mixed model). Multivariate data analysis (PLS-DA plots) of the liver metabolome revealed 3 different metabotypes (A = medium, B = minor, C = large alterations in liver metabolome profile between AP and PP status). Metabotype B animals were characterized by higher PP lipomobilization (stronger PP body condition decrease and higher blood bilirubin, fatty acids, gamma-glutamyltransferase, and triglyceride levels) and a higher occurrence of transition cow diseases, compared with the animals in metabotype C. Analysis of the feeding data showed that the period of metabotype B animals (calving in a distinct time frame) was characterized by a decreased grass silage quality. The PP liver metabolome of the metabotype C animals was characterized by higher concentrations of AA, acylcarnitines, lysoPC and sphingomyelins compared with metabotype B. For the metaphylactic treatment with BCC a dose-dependent effect was confirmed, differing between the metabotypes. In all matrices and metabotypes at various time points significant treatment effects were observed, with different profiles in clinical chemistry and as well in metabolomics data. The most clear-cut treatment effect was observed in metabotype B in the liver at 7 d PP, characterized by an increase in several acylcarnitines and phosphatidylcholines, indicating a more efficient influx and oxidation of fatty acids in mitochondria and thereby an increase in energy supply and more efficient triglyceride export in the liver. The results from the liver metabolomics analysis support the application of an indication-based metaphylactic treatment with BCC.