All,
Forgive me in advance if already addressed; I don't really know the phrase to search for an answer to my question, which is as follows:
Background: I work in commercial real estate (CRE) where I have access to numerous resources, both internal and external. For example, we use economic forecasting services, CRE sales and forecasting vendors, REIT-derived information and forecasting, internally maintained database of transactions, census data, geo-tagged news articles, etc. The goal of our PowerBI project has been to consolidate all of these resources within the PowerBI ecosystem (through table importation, API's, links to MS Lists, etc), which we have done. However, when you start adding to the already-robust underlying census tract data, millions of datapoints from a variety of different sources (some with data back to the 80's), PowerBI gets a little sluggish.
Question: So, this begs the question: Is there a general rule of thumb people use when creating a data model (or, as I refer to it, a single .pbix file) where it makes more sense to break it down into multiple files? This is obviously less ideal, as it requires flipping back and forth between models, but it's not horrible as we use MS Teams for daily workflows (i.e. you can have a couple of tabs linked to different models within teams, which isn't too cumbersome). As an example, instead of having economic forecasts, interest rates, census incomes, supply pipeline, sales comp transactions, historical market performance, geotagged news articles, etc....all in one file, or would it make sense/is it best practice to break each topic into it's own .pbix file?
I believe you can maybe even reference visuals in other data models if you need to as well? Might this be a better solution?
I really am aiming to increase the 'snappiness' of PowerBI for my colleagues who may or may not have powerful laptops.
TIA
