2024 Oct 28 11:25 AM - edited 2024 Oct 28 11:27 AM
Let's think of a very common scenario. You have
1. Sales value
2. Stock value
You want to analyse sales vs stock value of a product. You need to combine those two facts.
From what I see you have the following sitation
1. Analytic model does not support more than one fact model and I can see no reference that this feature is being worked on
2. Business builder is abandoned
3. Blending models in SAC is not supported and I can see no reference that this feature is being worked on
Am I missing something here?
Sure you can aggregate the data based and create another analytics model but then you lose a lot of the granularity that you might need. I honestly don't understand how such essential features are not supported when the competition has been supporting this for years!
I feel like a fool when business users ask me completly understanable questions when they want to see different fact measures aggregated per customer, product and so on and I have to tell them that this is not possible with this quite expensive product.
Request clarification before answering.
Hey Alexander,
You need to model your data. Like SQL WH. For your scenario you just create a UNION - or JOIN and on top of this View you create you Analytic Model.
Actually very easy, straight forward and used in classical WH for decades.
There is also a very nice blog to this topic: https://community.sap.com/t5/technology-blogs-by-sap/sap-datasphere-analytic-model-series-blog-post-...
I hope this helps.
Cheers
Martin
PS: Business Builder - I never used it, after Analytic Model was introduced. But it is definitely not obsolete, rather the functionalities are moved to Data Builder (like the Versioning/Release State)
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Martin,
well yes working with joins and unions is always possible but there is a reason why SAC allows blending of modells (just not for datasphere models) or why business builder alows more than one fact model: It is much easier and safes time to be able to "connect" two fact models one way or another.
I think everybody who had to combine two very different fact models knows that it requires a lot of work and it is easy to make mistakes. Especially joins and the possibility of unwanted multiple entries always looms.
User | Count |
---|---|
74 | |
29 | |
8 | |
8 | |
7 | |
6 | |
6 | |
5 | |
5 | |
5 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.