Just a short background...
I have already bought and read the Super Charge Your Data Warehouse book. I have been reading articles and watching videos about the Data Vault Modeling approach and methodology for over a month now as I shall use it to build a Raw Data Vault to be the core of the new EDW in my company.
By coincidence, there are several business transformations happening at the company (a merge, full replacement of several legacy systems with integrated suites, organization changes, ...) which are expected to be resident at least during the upcoming 2-3 years. Unfortunate and very challenging for an EDW initiative. That was one of the essential reasons that pushed me away from the classical Kimball-only approach looking for an alternative till I learned about the Data Vault (which also just happended by coincidence ).
Now the question:
In such circumstances, it is inevitable to face the business key changes, either through old-to-new 1:1 mapping, 1:M split or M:1 consolidation. As Data Vault modeling considers the business (natural) key to be immutable. Is there an official method or a best practice to handle such changes?
There is of course the horrible option of killing all old affected relationships using effectivity satellites and just keep old data as a historical view and start with fresh hubs/links/satellites. However, there should be an option to enable the users to report from the data using both the old and new keys. Have not anybody face this issue previously? Any thoughts?