This page has 0 threads | Add post

Blogs

Finding our blogs useful? Please consider helping us to share more!

Stop automatic grouping in Power BI Desktop
Read this blog to understand the risk of Power BI's automatic grouping of data, and to learn how to avoid getting caught out!

Posted by Sam Lowrie on 18 July 2018

Preventing automatic grouping in Power BI Desktop

For those new to aggregation in programmes other than Excel, it's worth noting what happens when using aggregates in visuals in Power BI. To demonstrate this, consider this table:

A simple table listing the number of products sold in each store.

The Problem

Everything looks fine at first glance, although Arndale Centre looks a bit of an outlier. Maybe it is in London? Let's check the towns these centres belong to:

Uh oh! Turns out that there are several centres called Arndale Centre - Power BI has grouped the sales to centres of the same name.

You can see what's going on more clearly if you count how many times each Centre appears:

Each centre has its own row in the table, so we can check if the same name appears multiple times.

Creating or using a unique centre identifer

The easiest way to fix this problem is to use a unique identifier for each centre within the table visual to split up centres with the same name. We already have a unique column called CentreID:

Make sure to un-aggregate the column by choosing Don't Summarize, otherwise Power BI will try to SUM or COUNT it.

If you don't have a nice ready-made column to hand, not to fear. We can add one pretty easily in Query Editor. First click this button:

The Edit Queries button on the Home tab.

We could do a similar thing by using the RAND or RANDBETWEEN functions in a calculated column.

Select the table you want to add the new unique column to (for us it is the Centre table). Then choose Index Column from the Add Column tab: