Continuous manufacturing of pharmaceuticals becomes more important due to greater efficiency, cost reduction and better quality assurance compared to batch production. For the continuous manufacturing process to be valid it is necessary to provide adequate methods for real- time process monitoring in accordance with the ICH Q2 (R1) guidelines. Raman spectroscopy is a non-invasive vibrational spectroscopic technique that enables rapid quantitative characterization of analytes in a sample. For this reason, a conventional Raman device equipped with a probe implemented in the production process is a promising way to monitor the active pharmaceutical ingredient (API) content in real-time. In this research we investigated the suitability of Raman spectroscopy as an in-process control (IPC) method in the production of liquid pharmaceuticals with 1% w/w or less API. Raman models were created for Meralys® spray containing 1 and 0.5 mg/ml xylometazoline hydrochloride (Xylo HCl) and for Aknet® solution containing 10 mg/ml clindamycin hydrochloride (Clinda HCl). Univariate integration and hard models as well as multivariate partial least squares (PLS) models were created. The integration models did not provide validation parameters within the requirements for any of the APIs. For Xylo HCl, PLS model gave best results with a correlation coefficient (R2) of 0.999, a prediction error (RMSEP) of 0.92% and a difference between the real and the predicted API content from -1.98 to +1.86%. For Clinda HCl, hard model gave the best results with an R2 of 0.997, an RMSEP of 0.38% and the difference ranging from -0.70 to 0.98%. None of the developed models met all the validation parameters. Conventional Raman spectroscopy is not a suitable method for analyzing API content in selected products and as such cannot be accepted as IPC in accordance with ICH Q2 (R1) guidelines.