Static Rank beast mode

dacorson

When creating multiple ranking tiles in ETL, the performance is significantly slowed and the dataflow takes hours to run, in comparison with less than 1 hour without ranking tiles. Therefore, I would like to be able to use beast modes to create static rankings. Currently, if you use the rank function in a beast mode, the rankings change dynamically when filtering on a particular brand (in my data this is for brand ranking). So, for example, if in the entire universe, brand X has a rank of 5, when I filter on brand X, the rank changes to 1 because it is the only brand in the visual context. So if I want to filter on 5 random brands to see where they rank, the beast mode recalculates the ranks as 1-5, but they may not truly be top 5 brands in the dataset. Since there are a large number of ranking scenarios, I don't want to have to add new ranking tiles to the ETL any time something is requested. It would be much easier to create a static rank beast mode (though I'm not sure how easy it is to create this code).

The fixed function, for example, adds the ability to create a static share of total by using the function in the denominator of the share calculation, so something like this for ranking would be a big improvement.