PowerBI changes filtering behaviour with one to many relationships

Data quality assessment is a task that I find I frequently need to do quickly and efficiently as I work with new data. Say I get two tables with ID’s with different attributes. It’s common to check how many of those ID’s are common, and how many of them might be missing in one table or the other. If the relationship was one to many, it used to be possible to create a table with ID’s from both tables, filter on blanks, and quickly see how many were missing. (The solution here is different if it’s many to many).

I’ve noticed that the “Show items with no data” has changed the way filtering works. Suppose I have two tables, fact and dim. Fact contains the values 3,4,5,5 and dim contains 1,2,3. They are related with a one (dim) to many (fact) relationship with a single cross filter direction.

The show items with no data option will show items that are in the dim but not in the fact. Namely; 1,2. That seems reasonable.

What I’ve noticed, is that you can no longer filter factID where it’s blank.  You would expect that to work, and give you dimID of {1,2},  but it does not.  Instead, it will give you everything in the dimension.

If this were a 1-1 relationship, it would work as described above.  You could filter blanks on either side and get the desired effect.  (The image below is how it behaves)

How the two tables appear under a one to one relationship

How the filter works under a one to one relationship

So if you’re looking to understand what is missing in the 1 to many relationship, the only way to do that is to create a measure in the dimension table, counting the number of factID’s.  Only then, can you filter where they are blank, and see what’s missing.

Fact count exists in the dimension table. The row where dimID is blank and FactCount=3 shows the three records that are in the fact table {4,5,5}, but not in the dimension table

Filtering where FactCount is blank gives us the dimID’s that are not in the fact. This is what we could not accomplish in using the two values and relationship in the third image of this post

 

So there you have it.  When you filter for blanks on the “many” of a one to many, you will be looking at all of the items in the dimension, instead of just the ones that don’t have a factID.   Here’s is the PBIX file if you want to download it and play.

Far better an approximate answer to the right question, which is often vague, than the exact answer to the wrong question, which can always be made precise. -John Tukey
The plural of anecdote is not data. - John Myles White

Recent Posts

RSS PowerBI blog

  • End-User Filtered Export capabilities now available February 13, 2019
    For the third straight week, we’re delivering on the item I’ve heard more end-user feedback around than any other we’d promised to deliver in the October blogpost.  At long last, you and your users will be able to export their Power BI reports to either PDF or PowerPoint and have their selections on the screen […]
  • Incremental refresh & query folding February 13, 2019
    As part of our strategy to converge enterprise and self-service BI on Power BI as a single platform, we announced the public preview of incremental refresh last summer. While there is still more work to be done to get it to general availability, we have seen strong uptake of incremental refresh. Incremental refresh is a […]
  • Microsoft a Leader in Gartner’s Magic Quadrant for Analytics and BI Platforms for 12 consecutive years February 13, 2019
      We’re very grateful to our customers, our community members, and our partners for making Power BI what it is today.   Thank you.   The Power BI Team       Get the 2019 Gartner’s Magic Quadrant for Analytics and Business Intelligence Platforms report* to learn more.             *This graphic […]