The aim of this study was to evaluate the impact of different inactivation and splitting procedures on influenza vaccine product composition, stability and recovery to support transfer of process technology. Four split and two whole inactivated virus (WIV) influenza vaccine bulks were produced and compared with respect to release criteria, stability of the bulk and haemagglutinin recovery. One clarified harvest of influenza H3N2 A/Uruguay virus prepared on 25.000 fertilized eggs was divided equally over six downstream processes. The main unit operation for purification was sucrose gradient zonal ultracentrifugation. The inactivation of the virus was performed with either formaldehyde in phosphate buffer or with beta-propiolactone in citrate buffer. For splitting of the viral products in presence of Tween®, either Triton™ X-100 or di-ethyl-ether was used. Removal of ether was established by centrifugation and evaporation, whereas removal of Triton-X100 was performed by hydrophobic interaction chromatography. All products were sterile filtered and subjected to a 5 months real time stability study. In all processes, major product losses were measured after sterile filtration; with larger losses for split virus than for WIV. The beta-propiolactone inactivation on average resulted in higher recoveries compared to processes using formaldehyde inactivation. Especially ether split formaldehyde product showed low recovery and least stability over a period of five months.