I need to convert the following beast mode to Group by

Absher_645 · 2025-06-13T13:48:35+00:00

There was an error rendering this rich post.

SUM(served) OVER (PARTITION BY Physical County,CABINET) / Sum(Passed) OVER (PARTITION BY Physical County,CABINET )*100

Find more posts tagged with

Accepted answers

All comments

Absher_645

This calculation uses both aggregated and non-aggregated columns. This may result in duplicate rows and unexpected behavior in charts. See this article for more information.

Absher_645

I am getting duplicates and unexpected behavior

ArborRose

Can you just put Physical County and Cabinet on a card and then add a formula (beast mode) as

(SUM(`served`) / SUM(`passed`)) * 100

Your group by is in essence the Physical County, Cabinet.

Absher_645

Tried that and the SUM was wrong before the math was good just inconsistent and duplicating

ArborRose

What about using FIXED? Something like

SUM(`served`) FIXED (BY `Physical County`, `CABINET`) / 

NULLIF(SUM(`passed`) FIXED (BY `Physical County`, `CABINET`), 0) * 100

Absher_645

that kinda worked but now I get

An issue has occurred during processing. We are unable to complete the request at this time - 'undefined'

and am unable to save the card

ArborRose

A bit difficult to visualize without an anonymized data example. I'm thinking there's an issue with NULL.