Enterprise Architecture (EA) models capture the fundamental elements of organizations and their relationships to serve documentation, analysis and planning purposes. As the elements and their relationships change over time, EA planning becomes increasingly complex. An analysis of existing methods shows that the complexity of dynamics is not sufficiently addressed. We argue that a sophisticated understanding of the complexity matter is prerequisite for EA planning method construction. As Chaos Theory (CT) is deployed in natural and social sciences—as well as in different contexts of IS research—to describe and understand the behavior of complex systems over time, we use properties of CT to assess the complexity of dynamics in EA planning and to derive requirements for EA planning methods. Our findings emphasize the importance of initial conditions of the architecture for EA planning and the need to harmonize planning granularities in order to achieve predictable results.