Over the past 2 decades, guidelines for the on-demand treatment of hereditary angioedema attacks have undergone significant evolution. Early treatment guidelines, such as the Canadian 2003 International Consensus Algorithm, often gated on-demand treatment by attack location and/or severity. Pivotal trials for on-demand injectable treatments (plasma-derived C1 esterase inhibitor, icatibant, ecallantide [United States only], and recombinant human C1 esterase inhibitor), which were approved in the United States and the European Union between 2008 and 2014, were designed accordingly. Subsequent post hoc analyses of clinical trial data alongside real-world evidence led to a paradigm shift. In 2013, the US Hereditary Angioedema Association guidelines recommended that all attacks, irrespective of location or severity, be considered for treatment as early as possible after onset to minimize morbidity and mortality. This approach remains the cornerstone of current treatment guidelines and has shaped the design of recent clinical trials, such as those for the investigational agents, oral plasma kallikrein inhibitor sebetralstat and oral bradykinin B2 receptor antagonist deucrictibant. This narrative review discusses the evolution of on-demand treatment guidelines, the clinical trial and real-world data that prompted significant revisions, and the subsequent changes to trial designs introduced to facilitate guideline compliance.