While there is extensive knowledge about how to design fiscal decentralization policies, considerably less is understood about how a decentralization program should be sequenced and implemented. Countries embarking on decentralization often struggle with decisions about the essential components of decentralization, including the order of an introduction of decentralization policies, the number of years necessary to bring a full program on line, and the components of the transition strategy. The authors argue that the sequencing of decentralization policies is an important determinant of its success. The consequences of a poorly sequenced decentralization program can range from minor delays and complications to ineffectiveness and subsequent failing support of decentralization efforts, macroeconomic instability, and fundamental failure in public sector delivery. At a minimum, the strategy of "making it up as we go" will not lead to the same structure of decentralization as will a planned strategy. The paper raises two questions: First, is there an optimal sequencing for decentralization policies and implementation? The answer is that there is, and that following these sequencing rules can reduce the costs and risks of implementing fiscal decentralization. Second, to what extent do countries follow these optimal sequencing rules? The answer is, in general, they do not. The gap between theory and practice is a result of the complexity of sequencing design, which discourages fiscal planners from implementing the full process. In addition, sequencing requires a sustained discipline and vision for its implementation, as well as overcoming pressures from political actors, especially in developing countries.