Decisions made by customers are often formed in a universe of possible decisions nested in a hierarchy and hierarchical choice models can be used to evaluate how changes to offerings will impact behavior. This method is often used in areas such as transportation modeling or television viewing, where there are a finite number of possible decisions clearly delineated in hierarchy and where early decisions impact later decisions. For example, the reason television viewing is simulated using this technique is that there are a number of linked decisions made by a viewer that leads to a program being viewed and the viewer makes these decisions in a particular order: "Am I busy, or can I watch TV?" Those that say yes, would then have to answer, "what shows do I like tonight." Then they have to answer, "Is that show on?" Yes, it's on, but in reruns. "Did I see the show already?" Yes, I did. "Do I watch it again?" No I don't. "Do I watch my next favorite show, do I check out this new program or do I do something else?" I try out this new program.
Clearly, hierarchical choice models can become quite complex at they are an excellent choice for simulating systems with many decisions points.