Feature Suggestion: Group by Secondary Domain

[Sean Larson]

Hello, I am building an application which allows users to compare outputs based on many scenarios of inputs meaning the same "domain" values of "name" will be repeated for each "Scenario".

This means I have a two values which identify a unique record, "Scenario" and "Name", see the screenshot

I can currently display the "Name" as the true domain along the hAxis, but I am currently hacking the second domain "Scenario" by adding it as an annotation of either the domain or the last series in a stacked column. 

This is not terrible, but the benchmark I use for display is Excel, see screenshot. I could achieve something similar through Javascript but think this is a good candidate for a feature addition.

As a side note, I loaded the "Scenario" into the datatable such that I could execute sort/filter commands.

Perhaps utilize the "columnRole" or accept a [ ] in the domain position?