Cocaine is the most widely used illicit drug, and its origin is always the focus of intense investigation aimed at identifying the trafficking routes. Since NMR represents a unique methodology for performing chemical identification and quantification, here it is proposed a strategy based on (1)H NMR spectral analysis in conjunction with multivariate analysis to identify the chemical "fingerprint" of cocaine samples, and to link cocaine samples based on this information. The most relevant spectral regions containing the fingerprint have been identified: δH 0.86-0.96, 1.50-1.56, 5.90-5.93, 6.48-6.52, 7.31-7.34, 7.61-7.63, 7.68-7.72ppm. The strategy has been applied on samples seized in different times and places in Naples (Italy). The chemical "fingerprint" depend on what plant they were extracted from, where it was cultivated, and which procedures were used for extraction and purification, thus adding significant information in the process toward identification of the trafficking routes for this drug.