Treemap bug: showing different data in filtered than in unfiltered view

[BramPeerlings]

Hi,

I'm using treemaps to show the number of occurances of particular events, grouped by countries and subsequently by cities. Not applying any filters to either countries or cities, the UK-set seemed to be missing data. That results in the figure posted here. (For some weird reason, this forum doesn't allow me to embed that picture... apologies for the slight inconvenience.)

If I now apply a filter to only show data from the UK, this image results. As you can see, large amounts of data from London and Manchester suddently appear - which were not present in the unfiltered view.

If I right-click the unfiltered graph and Show Data, the table shows Countries on the Y-axis (rows) and Cities on the X-axis (columns). That X-axis doesn't show all cities; the last one shown starts with a G. That would be consistent with the data as visualised (M and L would be after G).

That doesn't change the fact, however, that the (unfiltered) visualisation doesn't show what it's supposed to. As I have never seen any warning or error message popping up, I'm assuming this behaviour is either unknown (rendering it a bug, and raising serious doubts on quality control) or somehow by design (raising even more questions).

I'm hoping anybody here can point out to me what I might be doing wrong, or what I can do to resolve this issue. Furthermore, a reply from an official support person would be greatly appreciated, in order to see Microsoft's stance on this.

[BramPeerlings]

Thanks for your replies! I was expecting an e-mail notification, but that's apparently not how things work around here. Anyway, apologies for the delay in getting back to you - and thanks for the ideas and suggestions!

@amitchandak: At first I didn't see it, but looking carefully, I managed to uncover a light grey exclamation mark that indeed shows the 'Too many values' error. If anyone form the dev team is listening in on this, I suggest to make that exclamation stand out a little more. It's easy to miss (also because it only shows upon hover), whereas the results are quite dramatic - graphs showing data that leads to completely inaccurate conclusions.

I'll have a look at the workaround you proposed, amitchandak, to see whether that would work in our case. Thanks for sharing that idea, in any case.

@v-shex-msft:  As the issue is resolved, I don't think it's relevant to share the workbook with you, is it? It would be greatly appreciated, however, if you could pass my above message along to the developers. Even though it turns out not to be a bug, it seems a quite substantial UI/UX-issue to me.

[v-shex-msft]

HI @BramPeerlings,

If you can please share a pbix file with minimum data fields to test, it should help with troubleshooting.

Notice: remove sensitive data before share.

Regards,

Xiaoxin Sheng

[amitchandak]

Just check on the top right of a treemap is there an icon, which shows too many values(In the first condition). ! or some yellow icon. I faced wrong value issues when I had too may value passed to heat map.  I reduced the number of values passed and got the correct values.

Create a rank on cities and restrict to top 10 or 20

Score Rank = RANKX(all('Geo'[city]),CALCULATE(sum('Table'[value])),,DESC,Dense)

I create a rank for my details field (city) and then drag it to visual level filter and choose <10

Appreciate your Kudos. In case, this is the solution you are looking for, mark it as the Solution.
In case it does not help, please provide additional information and mark me with @
Thanks. My Recent Blogs -Decoding Direct Query - Time Intelligence, Winner Coloring on MAP, HR Analytics, Power BI Working with Non-Standard TimeAnd Comparing Data Across Date Ranges
Connect on Linkedin