While among researchers simulation models are consider to be useful tools, modelling has a badreputation in the industry. Many people feel it has very little connection with what happens in the field. Part of the problem is a misunderstanding of how models are developed and what sort of information we can expect to get from them. In this paper I will consider some of the steps we go through in building a model and what we can do with them. I will also give an outline of the CERCOT model currently under development at the Australian Cotton Research Institute.