Forum Discussion
Helper I"show items with no data" breaks the visual
Hi friends,
On of my coworkers is trying to make a simple table visual in Power BI Desktop that shows all objects in the system, even those without a 'last job date'. The table itself works fine untill she clicks on "show items with no data" and the table keeps loading endlessly. Even when it's supposed to show only 30 rows of data.
'last job date' is a measure. The data is saved in a tabular model. The relationship between the tables used is left/one direction/1-*.
I checked the tabular model server and it is having no issues and is not using much of it's storage and CPU.
Has anyone encountered this? Any ideas in how to fix it?
I suggested making a calculated column for 'last job date' as a last resort, but we prefer using the measure for that.
Thanks in advance!
Kind regards
Hi SokuTaIke ,
This issue sounds like it might be related to how Power BI handles measures in combination with the "Show items with no data" option. Since 'last job date' is a measure, Power BI tries to evaluate it for every row—even for those with no related data—which can cause performance issues or visuals to hang, especially if the relationships are complex or involve large datasets.The fact that the table loads endlessly, even for just 30 rows, suggests that the measure logic may be hitting a bottleneck when trying to compute over missing data. One workaround would be to ensure that the measure includes logic to handle blank or missing values more efficiently, using functions like COALESCE() or IF(ISBLANK(...)).
However, since you're using a tabular model and want to avoid creating a calculated column, another approach could be to create a disconnected table or a simplified supporting table that predefines the list of objects, then use that in the visual while referencing the measure separately. This way, you might be able to bypass the issue without altering the model structure significantly.
3 Replies
rohit1991Super User
Hi SokuTaIke
The issue likely comes from the fact that the visual relies on a measure that doesn’t handle blank results well when the related data isn’t there particularly for objects with no activity during the selected period. When that happens, Power BI struggles to materialize those rows, and the visual may go blank.
To fix this, by wrapping the measure in IF(ISBLANK([YourMeasure]), 0, [YourMeasure]). That forces the visual to treat blank rows as valid and display them as zeros instead of removing them entirely. This helps ensure that “Show items with no data” can do its job properly.
If performance is still an issue after this fix (especially with a large data model), you can also:
-
Move your time logic to a disconnected date table if it’s only used for filtering visuals.
-
Check whether your relationships are unnecessarily complex or many-to-many that can cause visuals to fail when blanks exist.
-
- rohit1991
Hi SokuTaIke ,
This issue sounds like it might be related to how Power BI handles measures in combination with the "Show items with no data" option. Since 'last job date' is a measure, Power BI tries to evaluate it for every row—even for those with no related data—which can cause performance issues or visuals to hang, especially if the relationships are complex or involve large datasets.The fact that the table loads endlessly, even for just 30 rows, suggests that the measure logic may be hitting a bottleneck when trying to compute over missing data. One workaround would be to ensure that the measure includes logic to handle blank or missing values more efficiently, using functions like COALESCE() or IF(ISBLANK(...)).
However, since you're using a tabular model and want to avoid creating a calculated column, another approach could be to create a disconnected table or a simplified supporting table that predefines the list of objects, then use that in the visual while referencing the measure separately. This way, you might be able to bypass the issue without altering the model structure significantly.
SokuTaIkeThank you! This was the problem. We added IF(ISBLANK()) to the measure and the problem was solved.