Grouping by substring selected in a beastmode

Hello,

I'm looking to group by the left two (or more, depending on where I'll end up going) characters of a column. I made a calculated field of left(`from`,2) that gives me the first 2 characters, but I can't group by this value. I have a LOT of possibilities in the from field, so I can't create the hundreds and hundreds of specific cases. I'd like to do this at the card level rather than ETL, but I would just have done a group by `from` statement in SQL to get the expected result.

Here's an image of my table to give you an idea:

Thanks!

Find more posts tagged with

Comments

KaLin

Can anyone help with this request?

JasonWebber

It would be nice to group by in beastmode to get this functionality. I wonder if there is a workaround...

You're in the table card, so the way you effectively GROUP BY is by selecting your aggregation level on the measure column. Every other dimension column is assumed to be part of the GROUP BY clause and your measure will be summed, counted, maxed, or whatever.

You'll just have to take out of the visualization any column you don't want to group by in the aggregation.

If that suggestion isn't on target, maybe mock up an example of what you're hoping to get, and post that up here.