The process of instructional design has parallels in other design disciplines. Software design is one that has experienced intense attention in the last 30 or so years, and many lessons learned there can be applied to ID. Using software design as a springboard, this concept paper seeks to propose a new approach to ID. It suggests that instructional design is almost always a Wicked Problem. The connection is formed between Wicked Problems as first described by Rittel and Webber in 1973, and the models in and processes of instructional design. The areas of social planning, organizational management and software design all posses some accepted and tested approaches to the solution of Wicked Problems. These will be described, and how they can be applied to ID will be explained. Finally, this paper will propose a meta-model for ID and explain how it can be used in the current context.