In ecology and other sciences, mathematical models can be constructed to help in prediction and in understanding observed phenomena. No model is perfect. A model is just a way to explain the observations available and squeeze them into a coherent generalized form.
In the business world, no model would be perfect to. If it was possible to construct a perfect model in business, then all we had to do is create a computer program, provide it with such model and feed it with data and it would predict perfectly the stock market and take perfect business decisions for us without our intervention. This is not only not present now, but is not possible to ever reach. The reasons this scenario is impossible to reach is because we can never create a perfect model and we cannot get all the data needed in the real world for a 'perfect' model to operate on. Nevertheless, we can still use models to help us understand the business world and guide our decisions. The feedback we get after applying the model can help us keep refining it. It is the lack of perfect predictability in the business world that makes all the fun. In business, experience and intuition are of very high value that cannot be compensated for with academic or ready models alone.