Forum Discussion
PowerBI - Comparison visualisation within a dataset using two filters of the same column
Hi All,
I am developing a tool on PowerBI and trying to create a comparison visualisation that allows me to apply two filters on the same dataset. I have created a sample visualisation below, which I was able to produce by duplicating the original table. However I would like a solution where I do not need to duplicate the table as this doubles the amount of data - and is causing space capacity issues.
Snapshot from the visualisation from PowerBI
As you can see from the visual i have selected two brands so that i can directly compare them across 'category'
Here is the sample dataset:-
And to make the above visualisation work i had to copy the table by creating a duplicate. Now this works, but effectively doubles the size of the data on the tool, which is not feasible for me. Can anybody recommend an alternative please? Example of data model i used for this visualisation:-
Thanks in advance!
Anonymous
You can try selecting the chart and going into Y axis. Here you can change Inner padding to, like, 5, so the bars don't look so scrawny. You can also ramp up the Minimum category width a bit for th same reason, although be careful as this will push your bars off the bounds of the chart if put up too high.
You can also Go to Legend and adjust where this sits to help with the balance of the visual.
Other than these little tweaks, there's not much else you can do. A bar chart is never going to look great with loads of data points on it, it's more for clear, low-volume, comparisons I think.
Maybe try switching to Clustered column chart and see if you find this a bit easier on the eye.
Pete
5 Replies
Hi Anonymous ,
Sorry if I'm oversimplifying this, but can't you just multiselect the brands in your dropdown list A so they both show on the chart?
Either ctrl+click on the second comparison brand on the dropdown, or select the slicer and go to Selection Controls and switch off 'Multi-select with CTRL' so you can just click on/off brands in the drop down.
Pete
Hi Anonymous
Add the brand into the legend, like this:
Pete