Cryo-electron tomography is a powerful tool to visualize heterogeneous samples, with one major application being structural characterization of pleomorphic viruses. In recent years, subtomogram averaging of viral glycoproteins has emerged as a method to directly visualize these crucial proteins on the surface of intact virions. One important target is the hemagglutinin (HA) glycoprotein of influenza virus, which densely covers the viral envelope and is responsible for influenza receptor binding and membrane fusion. While subtomogram averages of influenza HA have been reported, their resolutions have been limited due to the low signal-to-noise ratio inherent to cryoET as well as the manual effort required to analyze heterogenous influenza virions. Presented here is a cryoET analysis pipeline that integrates several software packages to analyze tomographic data of influenza virions efficiently and robustly. This protocol describes the structural determination of HA from influenza virions, through steps from initial motion correction to final model building. Following this pipeline, a HA reconstruction at 6.0 ? resolution was obtained from two cryoET datasets collected from the A/Puerto Rico/8/34 (PR8) influenza strain.