Analyses of extremist video messages typically focus on their discursive content. Using the case of ISIL (Islamic State of Iraq and the Levant), this study instead draws upon the emerging subfield of genosonic analysis to understand the allure of extremist videos, as well as the ineffectiveness of US video messages designed to ‘counter violent extremism’ (CVE). Through a genosonic analysis of three high-profile ISIL videos and five popular US State Department CVE videos, the study advances two concepts – sonorous communality and sonic unmaking – to help explain ISIL’s appeal. The lack of equivalent dimensions in US CVE videos renders them sonically sterile in comparison to those of ISIL. The implications of this analysis for scholarship and practice conducted at the intersection of media, war and conflict are discussed.