Business agility is key for companies across industry domains. As a primary mechanism for achieving agility, agile development practices have been successfully adopted by software organizations. However, while agile practices successfully support software agility, business agility also requires system agility. For companies in the embedded systems domain, this is particularly challenging since business agility is achieved only when all technologies in the system, i.e., mechanics, electronics, software, and artificial intelligence (AI) are subject to agile cycles. Although there is prominent research on business agility, industrial best practice shows that effective adoption of these frameworks is scarce. As a result, agility remains at primarily a software level. In this paper, we focus on companies in the embedded systems domain and ways in which these companies can increase system agility. The contribution of the paper is two-fold. First, we identify the limitations of contemporary agile frameworks. Second, we define a quantitative approach for determining the optimal release frequencies for the different technologies that are part of an embedded system.