Glycerol monolaurate (GML) is a food additive which is widely used. The products prepared by esterification of glycerol and lauric acid followed by molecular distillation have lauric acid, glycerol and other impurities. To detect the glycerol monolaurate, lauric acid and glycerol in the products is a very tedious process using traditional methods such as chemical titrimetry and gas chromatography. For fast quality detection of the products that has been analyzed on gas chromatography, Fourier transform infrared spectroscopy (FTIS) was used in this study. Principal component analysis (PCA) and back-propagation artificial neural networks (BP-PANN) were used to build the regression model. Then this model was used to detect the glycerol monolaurate, lauric acid and glycerol of samples in the validation set according to their spectra. Root mean square error prediction (RMSEP) and correlation coefficient (r) of prediction were used as the evaluation standards, the predicted results of GML was PRMSE＝3.6376, r＝0.9950, the predicted results of glycerol was PRMSE＝1.4764, r＝0.9795 and the predicted results of lauric acid was PRMSE＝1.2859, r＝0.9247. The results indicated that it was feasible to detect glycerol monolaurate, lauric acid and glycerol content using spectroscopic technique.