This paper challenges some widespread theoretical assumptions and practices in the study of populism and proposes a new multidimensional approach to generate and analyse data on this latent construct. Rather than focusing on categorising subjects as populists or not, it recommends reaching a better understanding of what populism is, the salience and relative weight of its attributes and how they interact creating an inner populist logic. Despite the increasing media and academic attention, historical discrepancies in how to conceptualise and operationalise populism have hindered cumulative progress in the literature. Initially most efforts were devoted to the study of specific movements, without a clear comparative angle, and the concept of populism was often conflated with that of nationalism. When the literature started to pay more attention to the analysis of the attributes associated with populism serious disagreements emerged concerning its true essence. Populism has been conceptualised as an ideology, a cynical strategy, a performative style and a discursive logic of articulation. The disputes between these competing interpretations have arguably slowed down the generation of comparative data. Although this article is meant to be a critique of the current state of the field and a call to make it pivot into a slightly different direction, it does not adopt an iconoclast stance and largely tries to reconcile the different existing research traditions – ideational, discursive, performative and strategic. It shows that their efforts are to a great extent complementary but mostly operating on different rungs of the ladder of abstraction. This paper argues that shifting from minimal definitions into a multidimensional approach may stimulate the generation of comparative data on a wider range of attributes and facilitate the identification of degrees and varieties within populism. This paper develops a new analytical framework which deconstructs populism into five dimensions: (1) depiction of the polity, (2) morality, (3) construction of society, (4) sovereignty and (5) leadership. These dimensions, that synthesise the most influential conceptualisations of populism, are empirically and theoretically interconnected and encompass ideational, discursive and performative attributes suggested in the literature. These dimensions are in turn composed of lower order attributes forming a multilayered network structure. This multidimensional framework provides a heuristic template that can be adapted and operationalised in diverse ways depending on the hypotheses, type of data and subjects of the analysis. Some examples of how to turn these dimensions into variables to capture supply‐ and demand‐side populism are introduced. Future empirical research could help map and better understand the network of interactions and intersections among these dimensions and attributes. This could be the key to settle some of the current conceptual debates about populism and its varieties. [ABSTRACT FROM AUTHOR]