From a design science research perspective, enterprise information systems (EIS) are understood as artifacts intended to support organizations in achieving certain goals. Proper EIS design needs not only to be based on solid general foundations (‘kernel theories’) and valid construction processes, but also should incorporate domain related experience and expertise. One important aspect is to understand which design goals and context factors have lead to which variations in existing solutions in the real world. Another aspect is to understand which design variations can be empirically related to which design goals, and to derive respective design actions. Using examples from enterprise architecture management and process performance management for illustration purposes, we show that existing variations of EIS solutions can be transparently explained and that innovative EIS solutions can be systematically constructed.