Tears are exceptionally reach sources of information on the health status of the eyes, as well as of the whole body functionality, due to the presence of a large variety of salts and organic components that can be altered by pathologies, eye's diseases and/or inflammatory processes. Surface enhanced Raman spectroscopy (SERS) provides an unique methods for analyzing low concentration organic fluids as tears. In this work, home made SERS substrates were prepared by using gold nanoparticles for investigating fluid samples excited by a HeNe laser (λ= 633 nm). The method was preliminary tested on Rhodamine 6G aqueous solutions at different concentrations, proving the possibility to sense substance concentration as low as the levels of the main organic components of the human tear. Tears were extracted from informed healthy patients by means of a not-invasive methods based on a direct collection of the tear fluid by means of a suitable micro-capillary tube. A clear SERS response was obtained for human tear samples. The SERS spectra were analyzed by using a “wavelet” based algorithm and a deconvolution procedure in terms of Lorentzian curves, in order to identify the contributions of biochemical components. The SERS spectra analysis allowed a preliminary insight on tear composition.