Hi @lps - A bit confused by your question as I don't have any SKUs in SYS12 - SKU Details module with "Chicago" as their Distribution Center. Is your concern that a user could select Chicago as a Distribution Center location in SYS12 and then SYS12 would be out of sync with SYS07?
David, yes, you are spot on with the concern. As data for both of these are imported, this leaves a potential for out-of-sync. I am aware that in a spoke Model, e.g. Supply Chain, subsets can be defined and thus eliminating the out-of-sync issue.
There seems to be a potential for out-of-sync in Data Hub model as these are imported in SYS07 and SYS12. Please refer to @einas.ibrahim 's reply. I would like to hear from those who are experienced if importing this way is a best practice and what are the reasons not to define Distribution Center as subsets in a Data Hub model?
Below is the activity for SYS12 from Level 2 - Sprint 1 - Data Hub
For the first point, it is best practice to build subsets in the spoke model and not in the datahub. It is best practice to import to lists from saved views and not other lists in the Datahub. So if you create the subset list in the datahub how will you access it in the spoke model according to best practice?
For the data integrity issue, remember this is the Datahub. No user will have access to it, so changes to the data are limited to imports and actions. If your imports are wrong - assigning Chicago to an SKU - then there is a bigger issue that should be addressed outside of Anaplan. According to the Anaplan Way, "Cleaning" the data is a process done by the customer and should be finished before model building starts.
Distribution centres that are marked as booleans can be used to create a subset of. As long as you are using that subset in the formatting of line items in SYS 12 module it is never going to throw you off.
You are worried that Chicago - which is not a distribution center (SYS07) - could be inputted as the distribution center to one of the SKUs in SYS12. This is a very valid concern, however, since these are System Modules, they are not intended to be accessible to the End User. If you want the user to add a new SKU and assign a distribution center for it or update the distribution center for an existing SKU then when you create that output module, you should format the Distribution Center Line item in that module as a list. This list is the Distribution Center subset list.
This is how you ensure the integration of the data in Anaplan.
The examples given were from the activities in the Data Hub model. Would this be considered a best practice approach? What could be the reason(s) not to define the Distribution Center as a subset in a Data Hub model?