The Heliospheric Imager (HI) instruments, aboard the twin STEREO satellites, have been used, with good results, in tracking Coronal Mass Ejections (CMEs) out into the inner heliosphere. There are now several publicly available catalogues of CMEs observed in the HI instruments, each of which provides the time-elongation profiles of the CME fronts that are identified by independent methods: identification by an expert (the RAL-HI event list), identification by citizen science (the Solar Stormwatch CME catalogue) and identification with an automated algorithm. The time-elongation profiles of the CME fronts provided by these catalogues can be used to estimate the kinematics of the CMEs and investigate the interaction of the CMEs with the solar wind. However, systematic differences exist between the time-elongation profiles provided by each catalogue, due to the different methods used to characterise CME fronts in the HI data. This investigation examines these differences by comparing the CME fronts identified by each method. More specifically, we aim to answer two primary questions: what are the quantitative differences between the time-elongation profiles and do these differences significantly affect estimated event properties? Furthermore, as both the Solar Stormwatch catalogue and the automated solar transient tracking algorithm presented here are both newer, less established methods, the comparison of these systems with the more commonly used approach employed by the RAL-HI event list also serves to as a partial validation of these new methods.