Our study is a collaborative dissertation paper that combines two different discourse analyses, textual and visual, based on a common theoretical background. The introduction guides the reader through the content of the study, at the same time offering a brief context of research. The aim of the paper is to address a gap that we identified in the study of mindfulness, namely a critical approach, from a media and communication perspective, of how this concept is represented in social media. Even though our research questions are developed separately in the analyses conducted independently, they can be reduced to three core questions: ‘How is the meaning of mindfulness constructed on Twitter?’, ‘Are there any power relations in the construction of discourse and if they exist, how do they shape the discourse?’, ‘How does the reproduction and circulation of discourse shape its meaning through intertextuality?’ For answering these questions existing research from psychology, sociology and business has been reviewed, with the mention that no relevant media and/or communication studies on mindfulness have been found. Therefore, our attempt to open a discussion in the field required a theoretical frame of analysis. For that we chose Michel Foucault’s discourse theory, adding observations on relations of power, and Stuart Hall’s theories of representation. The methodologies used for the two analyses are Fairclough’s and Rose’s approaches of applied discourse analysis. Critical Discourse Analysis (CDA) and Visual Discourse Analysis (VDA) are two detailed disseminations of qualitative data, conducted separately. Results show that there is a mainstream discourse that portrays mindfulness as a positive practice. This type of discourse might be invested with power, however our conclusions in this sense are restrained by the limitations of access to Twitter data. High intertextuality and low reliability on the scientific discourse further suggested in our case that the understandings of mindfulness are subject to change due to an advanced grade of interpretability among Twitter users.