This paper examines the question of decomposability versus complexity of teaching situations by presenting three case studies of MIS courses. Because all three courses were highly successful in their observed outcomes, the paper hypothesizes that if the attributes of effective course design are decomposable, one would expect to see a large number of common attributes emerge in the characteristics of all three courses. Instead, radical differences in course design and delivery are observed across all three courses. To explain how such different approaches can lead to successful outcomes, the paper draws upon the concept of a rugged fitness landscape (Kauffman, 1993), first introduced in evolutionary biology and later applied in informing science (Gill, 2008), wherein high levels of interactions between entity attributes necessarily lead to multiple fitness peaks. To support the proposition that the courses described exist on such a landscape, the courses (and the evolution of their designs) are examined for qualitative evidence of interactions between characteristics. Looking at four general areas--the instructor, the course content, the design/delivery method, and the students--evidence for the presence of interactions is observed. Thus, the three courses appear to confirm the hypothesis that the fitness of a particular course exists on a rugged landscape. The paper considers how landscape ruggedness may impact research in the area of course design. Informing science research has demonstrated, for example, that when entities on such a landscape individually attempt to maximize fitness, they tend to cluster on peaks. As a consequence, statistical approaches to explaining entity fitness, such as multiple regression analysis and structural equation modeling, may vastly exaggerate the significance of observed relationships (Gill & Sincich, 2008). The huge number of potential interactions between characteristics in even small models may also require huge numbers of observations to perform such tests (as is commonly the case in medicine). Thus, qualitative approaches to understanding the course fitness may become the only rigorous tools that can be applied. Arguably such research is likely to take a very different form--both in terms of length and descriptive content--than much of the past research published in the area of course design. (Contains 2 figures and 6 tables.)