Oh, I got it wrong.
So, I have input channels and group channels and additional output channels (main L+R and matrix channels). If I set up DCA’s with the same channels like I have done within the group channels, I can control the output level of a group with the dedicated DCA. I do not need to control the volume at the output of the group channel but can instead control volume at the input side instead. If I have linked the output of groups to several matrixes with different levels I keep those differences at the overall outputs when I change levels of input channels with DCA.
I found my solution. But I don’t want to take anything away from all the requests that have been written within this post. I took me a while to think it the other way round…