Use minimum parameters

When creating models to make it look realistic, various parameters that are probabilistic in nature are used. At the same time, there are a lot of such parameters. 

Each such parameter brings the probability of error and deviation of the model fact from the plan. 

When building a model, use the minimum number of really important parameters. At the same time, these parameters should be under your control. 

For example, information that the average customer in a competitor's market makes 10 payments per year is useless for your model if you don't understand how to ensure that number of payments. You only manage what you are responsible for.  

That's why you have to build a model based on the experience of the team.