"Inverted" Segmenter

[Syndicate_Admin]

Is it possible that, by default, a chart appears empty until one option (or more) is selected in a segmenter?

[Syndicate_Admin]

What I intended, which I suppose is not possible, is to use two different segmenters: one with the 10 most common substyles, and one with the rest. In this way, the user can quickly access the top 10 and, if they wish, can compare them with others on the list of the least common. And also that the graph appears empty at first to avoid an overload of lines.

[Syndicate_Admin]

Well, what you see in the photo, there is always a gray line corresponding to that sub-style, "Electropop", although the one that has been marked is a different one.

[Syndicate_Admin]

By the way, why do you press the style that you press always get a gray basal line corresponding to "Electropop"?

[Ritaf1983]

Sorry, I don't understand what are you trying to achieve.

[Syndicate_Admin]

Would it be possible to manage the graph with both segmenters at the same time? Mark one substyle in the first and another in the second and have both visible in the chart.

[Ritaf1983]

@Syndicate_Admin 
Please show what you mean.

[Syndicate_Admin]

https://www.dropbox.com/scl/fi/obwtn49ai01rwo06jzjte/filtro_subestilos.pbix?rlkey=o2x8or2bacnjmcv197...

[Ritaf1983]

Hi @Syndicate_Admin 
It didn't work because you used the same table for the graph and the slicer.
is filtered on the graph in this scenario is always 1, because of the legend filtering the values.
You should add dimension for the slicer with unique values.

Use it only for the slicer

On the graph use your original table.

update the dax for the filter to :

subestilo_filtrado = INT(ISFILTERED('Table for slicer'[nombre_subestilo]))

The pbix is attached

If this post helps, then please consider Accepting it as the solution to help the other members find it more quickly

[Syndicate_Admin]

It gives me this error message when trying to attach the .pbix:

[Ritaf1983]

Hi @Syndicate_Admin 
Save the pbix in one drive or something like that and share the link

[Syndicate_Admin]

Yes, I see that it works, and that in the YouTube video as well. The problem is that in my data model it doesn't, and I don't understand why.

[Ritaf1983]

Hi @Syndicate_Admin 
Please share a link to your dummy pbix.
I don't think that it is possible to understand what is going wrong without looking at it.

[Syndicate_Admin]

What I'm trying to do is a bit more complicated, but I've replicated the video example as a test and it doesn't work for me:

The table I'm starting from:

The measure that is used as a filter:

The result of filtering:

As you can see, all the values keep appearing.

[Ritaf1983]

Hi @Syndicate_Admin 
Yes, you can create a flag measure for selections and filter the graphs by it.

Please refer to the tutorial:

https://www.youtube.com/watch?v=sXn-QZqLD-8

If this post helps, then please consider Accepting it as the solution to help the other members find it more quickly

[Ritaf1983 · Accepted solution]

Hi @Syndicate_Admin 
Please refer to my linked PBIX.

If this post helps, then please consider Accepting it as the solution to help the other members find it more quickly