The relationship between complexity and usefulness can be captured by a U-shaped curve. This comment explores that relationship. Complexity may be useful for one of the main aims of developmental psychology (causal inference) but not for another (description of developmental phenomena). Currently, developmentalists conduct complex analyses that are not useful in pursuing either aim: The analyses are too complex to produce good description, and the complexity is not employed in a manner that facilitates causal inference. Further complicating matters is that the complexity is often not made explicit, as the model specification is not mathematical. In many cases the analyses involve data that are not representative of a recognizable population and/or were sampled in ways that involve the processes of interest. Complex analyses of such data often plumb the depths of usefulness. The key to better analyses is to align the complexity of analyses with the research questions of interest. In some cases, doing so will mean simplifying the analyses to produce better description. In others, it will mean reducing some forms of complexity (e.g., involving measurement) and better aligning analytical complexity with the complexity of the underlying processes. This comment concludes with 6 questions authors can ask themselves in planning their analyses to maximize the usefulness of the results.