Can you do groupby and aggregation inside dataset

I’m trying to implement a requirement in QuickSight but I can’t do that in Analysis (since QS does not support conditional functions with PRE_AGG)

I thought of a way to do that, but it requires that I aggregate the data inside dataset. If I add calculated field in dataset and use aggregation function (like sum). It does not generate a preview. The number of rows also remain the same, meaning it did not aggregate.
Even if I just remove some fields from existing dataset (and do not add any calculated field), yet number of rows remain same. Since I have repeating rows in my dataset, I thought it should atleast aggregate but it doesn’t seem to be happening this way, since again number of rows in final dataset are same as initial.

Is there anyway to do group by in dataset ?


If you use custom SQL you can aggregate while building the dataset.


1 Like

ah ok, thanks…though i wished there was an easier way as I’m already using Custom SQL. I wanted to do more processing on top of it.