How do you handle tables that are both facts and dimensions.
Example: customer.. My customer dimension is beginning to turn into a fact table. I have a request for # of drivers with suspended licenses. So a count of customers in my current dimension table. A customer is a driver which has a license. A license can be in different status's.. valid, suspended, etc...
My primary fact table is at a transaction grain. So I can see # of transactions by customer. When I add the license status field to my dimension, I could potentially pull # of transactions by a suspended license type. This analysis doesn't really make sense. I would want to add a license type key in my customer dimension table and create a license type dimension. This would snowflake the schema.