The production process, such as fermentation and purification etc., can significantly affect the relative ratio of matrix components in a multi-component antibiotic. The ratio of components can be varied in different products. This status causes a difficulty to assure the homogeneity and consistency between reference standards and test samples in potency determination, which hinders the results judgment and accuracy of a routine microbiological assay. In the current study, a multi-component antibiotic, teicoplanin, was selected as a model to explore the relationship between the ratio of matrix component and antibiotics potency. Single-component samples, TA3-1, TA2-1, and mixed-component samples, TA2-2.3, TA2-4.5, of teicoplanin were prepared and purified. Dose-response relationship of each sample has been determined by HPLC and microbiological assay, respectively. The accuracy of the potency result was guaranteed by choosing a test organism with the same sensitivity to each component of teicoplanin when there were differences existing in the ratio of components between the reference standard and the test sample. The experimental methods in current specifications can be replaced with the new potency determination method, which can provide a more realistic reflection of the biological activity of the product.