Best Of

Hey TMonty,

You will need to create a conditional column:

case when `timestamp` = max(`timestamp`) over() then 'Yes' else 'No' end

However, this will require that you have a timestamp column.

If you do not have a timestamp column, you can make one by appending the date and time columns and converting to timestamp format.

I have a beast mode that is comparing actual performance to a target (by categories) by taking the sum of actuals - sum of targets over an arbitrary date range, defined by user input into variables on the dashboard page. This works great as a quick filter control for an individual card, but when I make a filter card using that beast mode and put it on a dashboard with the cards it was able to filter as a quick filter, it doesn't apply to any of the cards on the dashboard.

I'm assuming an aggregate function used as a filter needs the context of the specific card it applies to to determine which rows it should filter. But I thought I'd ask in case someone knows where I'm going wrong or knows how to work around this. I don't think I can determine performance in the ETL because of the variable date range option on the dashboard (They can select any start/end date in the last 2 years).

Thanks!

EDIT: First comment made me realize I should have mentioned the beast mode is saved to the dataset! Thanks Colemen.

EDIT 2: Problem was user error; I was using FIXED functions in the sums which I failed to revert after realizing how to get the results I needed without them.

Have you saved the beastmode to the dataset?