This article proposes a critical approach to implementation studies and seeks to challenge the current uniform facade of the concept of "implementation." Inspired by Latour's actor-network theory, it argues that implementation can be perceived as an assemblage of constantly shifting interactions. Thus, instead of seeking to identify the factors that affect the implementation process, as is the focus of most implementation research, it proposes seeking to understand what goes on in implementation, thereby moving away from looking for variables that influence implementation to instead discuss implementation analysis and implementation types. This shift in focus, which has received limited attention in the literature, holds potential for both theory and practice. For theory, it introduces an understanding of implementation as an independent concept, composed of multilayered components that can be categorized into different types, in different contexts and at different times. And for practice, as well as for theory, it opens up another route for implementation analysis and design, which has not been articulated until now.