Table with no Metrics (Show Dimension Label and Value)

I have a need to show a table with no metrics. I want two columns. 1 with the Dimension Label and 1 with the Value. I've worked for hours trying different ways to get this result and nothing.


Here's what I get (Value only, with no "label"):

Image title



Here's what I want (used the DAP for easy display of the lables i mean):Image title

This shouldnt' be hard, but I'm too dumb to figure it out. I don't want to make seperate labels that "float" outside the table, if possible.


Any help would be appreciated!

I have tried transposing this, which would be great, but i end up with the "What i get" example above.

Image title


Devon,


Question is the Unknown a Null value and how is the metric set handeling Null Values? Would changing that give you the result you are wanting?

Yes, Unknown is Null here. No, it doesn't help. That's just the data I had up when I screen grabbed. I tried with only fields with "real" data in other iterations to no avail.

Thanks for the reply!

Ah ok now I get what you are going after. Sorry had the wrong idea.

I am going to assume when you say no metric you mean no measures, just Dimensions.


First thing this would only work for a single vaule set, there not going to be any way to show more than one "Location and the other 4 measures because this limits it to 2 columns.

If that all you have then I think using Lables, Data Lables, Rectangles and lines is the way to go (only if you realy want it to look like a table).


I have some Dimension only Tables but none that only have one set a values. The Dimension is going to be the column Header for that value I am not sure if you can make that lable look like/act a value. You can make the vluse a Header by putting it as a Column and not a row.


The table is just expecting there to be more than one set of values for each Dimension so it not going to bend to your will on this I dont think.


What you got when you Transposed it is the best you will end up with, it is just horizontal not vertical. That one you can have multiple Rows of data and with what you want where are the multiple rows of values going to go? Even if you are visualizing the next set woudl be the 4rd column and so on. I am not sure that would be a good readable table. I woudl rather see horizontal rows than vertical ones.

Hello Devon,


Transposing the metric set won’t pass the headers over so you will have to go into the data cube and perform an Unpivot in order to get what you want. The way to do this is to add Unpivot transform in your data cube. Here are some instructions on achieving this:


  • Go to the data cube powering your visualization, click the wire in between your data source and process result, Toolbar -> Insert Other -> Unpivot
  • Select all the columns in your data source so they can be unpivoted into separate columns
  • The column headers will be grouped in label and the corresponding data in the value column


  • Here is the result:


  • You can toggle between the “Display With Row Headers” and “Display As Flat Table” in the top right corner of the toolbar (when editing metric set fullscreen)


Please let me know if you have any questions about this.

Thanks! Unpivot doesn't work for us as it's not the full dataset and needs to be in the original format for other purposes.

I solved the issue by laying out the table horizontally, which isn't optimal.

An option to diplay Column names would be super useful in this case.

If you don’t want to effect the entire dataset with the unpivot, you could create a new cube with the same data or even with the current cube as the data source.


You can also try adding custom group headers to your table to overwrite the existing header. This will give you customization functionality of the header text, you can then hide the current headers so only the customized headers are displayed. Here is an article for a reference.