This study analyzes the modes of signs in the music video “We Can’t Be Friends” by Ariana Grande using Daniel Chandler’s theory (2007). This study aims to explain how modes of signs function to construct meaning and convey emotional depth within the music video, as well as to demonstrate how semiotic theory can help audiences understand messages related to personal struggles, anxiety, loneliness, and mental health issues. The data source of this research is the music video itself, while the data consist of visual elements such as images, settings, gestures, facial expressions, and objects found in the scenes. A qualitative descriptive method is applied by recording and repeatedly watching the video, capturing scenes that contain sign elements, and classifying each sign according to its mode. The analysis is supported by the theory of visual grammar proposed by Kress and van Leeuwen (2006), which examines representational, interactive, and compositional meanings to explain how visual elements construct meaning. The modes of sign found are iconic mode, indexical mode, and symbolic mode. Through these three modes, visual elements in the music video function as signs that help viewers interpret emotional experiences and narrative development. These signs show a direct cause-and-effect relationship and serve as evidence of emotional reactions, physical conditions, and ongoing processes within the narrative. Regarding the second research problem about the types of meanings associated with each mode of sign, the analysis shows that there are three types of meaning found, namely representational meaning, interactive meaning, and compositional meaning.