Post-failure landslide change detection is crucial for mitigation strategies. However, the methods used to investigate this issue all involve a tough workflow, and the free access Sentinel-2 satellite is underutilized. In this study, we use ten Sentinel-2 optical images to explore the effectiveness of using these images to detect post-landslide changes in the Huangnibazi landslide failure using an easy workflow. We found that the landslide can be qualitatively divided into a startup and acceleration stage, a front and lateral edge expansion stage, and a stabilization stage using time-series true color images. After the normalized difference vegetation index (NDVI) was calculated to identify landslide scars, which were validated using the unmanned aerial vehicle (UAV) orthoimages, we found that the same three change processes identified were also reflected by the landslide scar count change analysis in a quantitative way. Based on the three different stages, a red-green-blue (RGB) composite of the NDVI images was constructed and was found to reflect the different change period of the right and left landslide edges. Most importantly, the changes within a pixel unit were detected using an NDVI RGB composite with cold colors representing a retrogressive landslide mode. All of these findings indicate that the huge potential of the use of Sentinel-2 images in similar applications.