Перейти к содержимому

48 - Playing with Summarize and Groupby (Power Pivot)

WesleySon

0:00 / 0:00

48 - Playing with Summarize and Groupby (Power Pivot)

690 просмотров · 2 года назад
WesleySon
307 подписчиков
690 просмотров · 2 года назад
I wanted to play around with the Summarize and Groupby functions in DAX. Both do the same thing - they group a table. They work differently under the hood, and are used in different cases. The summary is: Use Addcolumns(Summarize()) when you work with columns that already exist in your table - use Groupby when you work with temporary new columns. Another thing I keep stressing in the video is: Be clear as to what questions you're trying to answer. Do you want to see all rows with sales above X? Only the salespeople with sales above X? Do you want to know how many rows were above and below X? Communicate clearly! This explains clustering aka why you don't use Summarize on calculated columns https://www.sqlbi.com/articles/all-th... This explains Currentgroup, and is also where I got the Countx(Currentgroup(), 1) from. https://dax.guide/currentgroup/ 00:00 Intro 00:24 Query 1 - Rows of sales above 1k 00:57 Summarize can add columns. Don't 01:48 Addcolumns 02:43 Filtering a query 04:01 Query 2 - Salespeople who sold 1k+ in at least 1 month 05:29 Summarize vs Groupby 06:09 Using Groupby instead 07:26 Selectcolumns to only return X columns 08:08 Query 3 - Rows above/below a threshold 10:21 Currentgroup 11:48 Playing with the threshold