Recent contributions to information systems theory suggest the primary role of a firm’s IT architecture is to facilitate agility, ensuring the continued alignment of a firm’s capabilities with a changing business environment. Despite advances in our understanding of the role of IT however, we lack robust empirical data on the precise mechanisms by which IT architecture impacts agility. As a result, it is difficult to identify specific managerial actions that can improve a firm’s architecture, and thereby lower the cost of responding to changing future demands.We address this gap in the literature by exploring the relationship between IT architecture and IT agility at the level of the individual components in the architecture. In particular, we draw upon modular systems theory to develop and test a series of propositions about the impact of component coupling on the cost of IT change. Our methods make use of “Design Structure Matrices” (DSMs), a novel network analysis technique that addresses the limitations of prior work in this area. DSMs allow us to i) identify discrete layers in the IT architecture associated with different types of components, and; ii) capture data on the dependencies between components; hence to assess their relative levels of coupling and position within the architecture. We test our propositions using empirical data from a large pharmaceutical firm, encompassing 407 architectural components and 1,157 dependencies between them. We show that different components occupy different positions in the IT architecture and possess different levels of coupling. We find measures of coupling predict IT agility – defined as the cost of making changes to software applications. The measure of coupling that best predicts agility is one that captures all direct and indirect connections between components (i.e., it captures the potential for changes to propagate via all possible paths between components). Our results reveal that different business groups within the firm face different costs of adaptation, deepening our understanding of the roots of organizational agility.