In the coming years, EUMETSAT's Meteosat Third Generation – Sounding (MTG-S) satellites will be launched with an instrument including valuable features on board. The MTG Infrared Sounder (MTG-IRS) will represent a major innovation for the monitoring of the chemical state of the atmosphere, since, at present, observations of these parameters mainly come from in situ measurements (geographically uneven) and from instruments on board polar-orbiting satellites (highly dependent on the scanning line of the satellite itself, which is limited, over a specific geographical area, to very few times per day). MTG-IRS will present a great deal of potential in the area of detecting different atmospheric species and will have the advantage of being based on a geostationary platform and acquiring data with a high temporal frequency (every 30 min over Europe), which makes it easier to track the transport of the species of interest. The present work aims to evaluate the potential impact, over a regional domain over Europe, of the assimilation of MTG-IRS radiances within a chemical transport model (CTM), Modèle de Chimie Atmosphérique de Grande Echelle (MOCAGE), operated by Météo-France. Since MTG-IRS is not yet in orbit, observations have been simulated using the observing system simulation experiment (OSSE) approach. Of the species to which MTG-IRS will be sensitive, the one treated in this study was ozone. The results obtained indicate that the assimilation of synthetic radiances of MTG-IRS always has a positive impact on the ozone analysis from MOCAGE. The relative average difference compared to the nature run (NR) in the ozone total columns improves from −30 % (no assimilation) to almost zero when MTG-IRS observations are available over the domain. Also remarkable is the reduction in the standard deviation of the difference with respect to the NR, which, in the area where MTG-IRS radiances are assimilated, reaches its lowest values (∼ 1.8 DU). When considering tropospheric columns, the improvement is also significant, from 15 %–20 % (no assimilation) down to 3 %. The error in the differences compared to the NR is lower than for total columns (minima ∼ 0.3 DU), due also to the lower concentrations of the tropospheric ozone field. Overall, the impact of assimilation is considerable over the whole vertical column: vertical variations are noticeably improved compared to what is obtained when no assimilation is performed (up to 25 % better).