A reliable identification of cells on the basis of their surface markers is of great interest for diagnostic and therapeutic applications. We present a multiplexed labeling and detection strategy that is applied to four microparticle populations, each mimicking cellular or bacterial samples with varying surface concentrations of up to four epitopes, using four distinct biotags that are meant to be used in conjunction with surface enhanced Raman spectroscopy (SERS) instead of fluorescence, together with microfluidics. Four populations of 6 μm polystyrene beads were incubated with different mixtures, "cocktails" of four SERS biotags (SBTs), simulating the approach that one would follow when seeking to identify multiple biomarkers encountered in biological applications. Populations were flowed in a microfluidic flow-focusing device and the SERS signal from individual beads was acquired during continuous flow. The spectrally rich SERS spectra enabled us to separate confidently the populations by utilizing principal component analysis (PCA). Also, using classical least squares (CLS), we were able to calculate the contributions of each SBT to the overall signal in each of the populations, and showed that the relative SBT contributions are consistent with the nominal percentage of each marker originally designed into that bead population, by functionalizing it with a given SBT cocktail. Our results demonstrate the multiplexing capability of SBTs in potential applications such as immunophenotyping.