ABSTRACT The personalisation of politics has been widely studied in the past few decades, especially in Western Europe, from several viewpoints, from the electoral to the party-centered one, from the governmental to the media-related one. Nonetheless, very few contributions have focused on the determinants of personalization, even if this latter process might be linked to more general – and much relevant –societal, institutional, or party-related phenomena. In this article, we aim to fill this gap by proposing a novel framework to explain the evolution of the impact of the personalization of politics – i.e., the fact that personalization has had a stronger or weaker influence – in Western Europe between the mid-1990s and the mid-2010s. We depart from cross-time and cross-country data collected in the first expert survey on the personalization of politics (the PoPES) and investigate the effect of three sets of possible determinants: societal changes (the decrease in the strength of the class cleavage or the diffusion of TV), institutional factors (the presence of presidential elections or personal-vote-driving electoral systems’ features), and party-related changes (the organizational decline of Western European parties). Regression analyses show that almost none of these determinants have influenced personalization's impact in Western Europe.