I have an issue with SAC changing the order of operations/aggregation the moment I bring in a linked model into the story. Any help / advice is greatly appreciated!
I started with creating a story with a single model and created 3 calculations (see screenshot below):
It is working fine and I've checked that the output for Value Z is exactly what I expect to see at different levels of aggregation / detail, i.e. area - region - national.
See attached screenshot (dummy data), Region 1 consists of 5 Areas - working with the single 'master' model, I get what I expect at each area level and region level, i.e. Value Z = 21% = (sum(X) / sum(Y)). Great.
But then I need to link it to another model, link type All Primary Data - the reason for linking is to bring additional dimensions (no measures) that exist in the second model, by linking using a common ID (I've also selected to link on ID, see below).
However, what I get with the linked models is problematic but only at Region / aggregated level:
But when I remove the linked model, the value goes back to what it should be at region, which is 21%.
So we know it's got to do with the linking.
My question is.... has anyone seen this behaviour, where the order of operation / calculation just changes when you have linked models in the story? What causes the inconsistency?
Changing the link type won't do anything as we're not acquiring any measure from the linked model, and I don't think it's got to do with the link type anyway, as calculated measures at ROW / Area level return correct values it's telling me they match fine, it's just that the order of operation i.e. the way SAC calculates Value Z % suddenly changes the moment I bring in another model.
Why does it not hold the same principle as when working with a single model?
Many thanks in advance for your help / advice / solution!