Event-driven architecture has been widely adopted in the software industry, emerging as an alternative to modular development to support rapid adaptations of constantly evolving systems. However, little is known about the effects of event-driven architecture on performance, stability, and software monitoring, among others. Consequently, professionals end up adopting it without any empirical evidence about its impact. Even worse, the current literature lacks studies that point to which emerging research directions need to be explored. This article proposes an agenda for future research based on the scarcity of literature in the field of event-oriented architecture. This agenda was derived from a literature review and a case study carried out, as well as from the authors' experience. Eight main topics were explored in this work: performance analysis, empirical studies, architectural stability, challenges to adopting, monitoring event streams, effects on software performance, broader challenges for adoption, and better monitoring of event-driven architecture. The findings reported help the researchers and developers in prioritizing the critical difficulties for uncovering the hidden potential of event-driven architecture. Finally, this article seeks to help researchers and professionals by proposing an agenda as a starting point for their research.