We constantly get into a situation where we are squeezed between data and dashboards that need variable granularity.
LOD calculations are a great way to resolve the issue but as the LODs only execute at run-time we have a performance hit for large volumes of data.
It would be of great value if we can have an option of materialized LOD.
While searching for similar ideas, I came across a post from Andrée Roos that explains the exact same idea beautifully. I have nothing more to add to his post, so I have just copied it here.
Idea explained in his Post:
The problem now is that fixed calculations are affected by Context Filters and Datasource Filters, which sometimes can be a good thing.
Another type of fixed calculation should be added, perhaps by the name MAT, where the calculation aren't affected by these filters and the calculation therefore can be materialized.