The goal of this work is to understand the phenomenon behind the poor injection repeatability oftenobserved with protein biopharmaceuticals. All the measurements were carried out in size exclusionchromatography (SEC) mode, using fifteen commercially available therapeutic monoclonal antibodies(mAbs). First of all, we proved that the variation of peak areas between consecutive injections was muchmore critical with highly concentrated mAb samples (up to 80 mg/mL), while the SEC measurementsof commercial mAbs having concentrations below 5 mg/mL were reliable. Second, we emphasized thatthe shapes, volumes and materials of the sample vial insert also contribute to the change of peak areasobserved during consecutive injections. In this study, six different insert models were studied and themost critical were the ones possessing the narrowest conical shape at the bottom. Furthermore, thehomogenization of samples (with pipette mixing rather than vortex) prior to analysis was of great interestand allows a significant improvement in injection repeatability. Finally, because the on-wall (on-insert)adsorption of mAbs reaches its equilibrium in around 50 min, it is advised to add a specified residencetime prior to injection to achieve repeatable injection.