I guess I don’t understand how a data source be marked as primary or secondary, and who decides which is which.
I will illustrate my question with the following scenario in an area where I have some understanding: identifying billing verification (BV) discrepancies. Sometimes best practice is to model the tariff as defined at the physical contract with the end customer, and model the tariff as defined in the billing system, and model the tariff as defined on the website of the operator, and model the tariff as filed with the legal department and more… Now we can take these variants and conduct a parallel run and see if there are any discrepancies between the various dialects of what should be single version tariff spec.
How would you mark each version? Which is prime and which is secondary? IMHO for proper BV you want to model all, regardless. The end result of this exercise should be a single tariff definition company wide.
