Rapid, non-invasive methods for the detection and identification of pathogens are becoming very important in the field of medicine and defence. There is a constant need for technologies that can rapidly detect and identify the presence of airborne, foodborne or waterborne pathogens and toxins with high precision. Conventional methods employed by microbiologists include phenotypic and genotypic methods. The most commonly used biochemical methods include enzymatic activity, gas production and compound metabolism for identification of bacteria. Most of the methods mentioned above are time consuming and require extensive sample preparation. Raman spectroscopy is a well established molecular spectroscopic technique that measures bond vibrations to decode molecular structure and chemical composition of samples. In the realm of biochemical analysis it is essential that a technique is non-invasive, preferably label-free and non-destructive. These unique features make Raman spectroscopy and its variants an important tool for identification of samples. However, biological samples are very complex as even a single cell or a bacterium is composed of a number of biomolecules such as proteins, lipids, carbohydrates and nucleic acids. Raman spectroscopic analysis yield spectral fingerprint unique to the biomolecules. Therefore, these spectral markers can be used for tracking diseases, studying effectiveness of drugs in cells and tissues, identification of pathogens and many other biological processes. Furthermore, accuracy of classification and prediction can be enhanced and automated by combining this technique with chemometric analysis. Sensitivity of detection can be improved employing SERS based approach. Raman spectroscopy based methods have gained popularity in the last few decades due to rapid development in instrumentation that has led to enhanced sensitivity and resolution. However, it has been a challenging issue so far to differentiate and detect pathogenic strains from non-pathogens of the same species. In this manuscript, the work initiated towards identification of pathogens has been discussed. Eight strains of food pathogens were used as a model system and was classified based on PC-LDA using the Raman signals. The classification accuracy was 100%. The potential and limits of Raman spectroscopy based technology for detection of pathogens in real environment has been discussed.