Traditional methods for bacterial identification include Gram staining, culturing, and biochemical assays for phenotypic characterization of the causative organism. These methods can be time-consuming because they require in vitro cultivation of the microorganisms. Recently, however, it has become possible to obtain chemical profiles for lipids, peptides, and proteins that are present in an intact organism, particularly now that new developments have been made for the efficient ionization of biomolecules. MS has therefore become the state-of-the-art technology for microorganism identification in microbiological clinical diagnosis. Here, we introduce an innovative sample preparation method for nonculture-based identification of bacteria in milk. The technique detects characteristic profiles of intact proteins (mostly ribosomal) with the recently introduced MALDI SepsityperTM Kit followed by MALDI-MS. In combination with a dedicated bioinformatics software tool for databank matching, the method allows for almost real-time and reliable genus and species identification. We demonstrate the sensitivity of this protocol by experimentally contaminating pasteurized and homogenized whole milk samples with bacterial loads of 103–108 colony-forming units (cfu) of laboratory strains of Escherichia coli, Enterococcus faecalis, and Staphylococcus aureus. For milk samples contaminated with a lower bacterial load (104 cfu mL−1), bacterial identification could be performed after initial incubation at 37°C for 4 h. The sensitivity of the method may be influenced by the bacterial species and count, and therefore, it must be optimized for the specific application. The proposed use of protein markers for nonculture-based bacterial identification allows for high-throughput detection of pathogens present in milk samples. This method could therefore be useful in the veterinary practice and in the dairy industry, such as for the diagnosis of subclinical mastitis and for the sanitary monitoring of raw and processed milk products.