A feasibility study with 20 voluntarily participating veterinary surgeries was carried out in order to test, if the consumption of antibiotics in livestock can be determined systematically. Information about the statutory documents on the application of drugs of the participating surgeries were entered in a central database and analysed systematically. Surgeries that treat only livestock used significantly more antibiotics (number of treatment units per veterinarian) per veterinarian than surgeries that treat also small animals. The comparison of small and large surgeries showed that veterinarians in small surgeries treated fewer pigs and more cattle than their colleagues in large surgeries (number of treatment units per veterinarian). All in all, tetracyclines counted for more than 50% of all substances used (regarding the amount), followed by beta-lactams (25%) and sulfonamides incl. trimethoprim (11%). In poultry, polypeptides and beta-lactams were used most frequently. While cephalosporines were used only in cattle in a noteworthy frequency, fluoroquinolones were applied to poultry in almost 12% of all applications (application of one substance to one animal at one day). In total, it was shown, that harmonized documentation of consumption of antibiotics is feasible, but the relation of antibiotics to the treated population is problematic which has to be considered in the future, The number of applications is more suitable to assess the antibiotic use than the amount in kg, because the latter is dependent of the dosage. The impact of highly dosed substances like e.g. tetracyclines is overestimated by regarding the amount, while substances with low dosages are underestimated.