Signal path and setting input levels a the various stages

[Nos402]

So here's what I'm currently using to record a podcast with a friend:

mic > mixer (Compressor/limiter/expander as an insert effect on each mic channel) > Komplete Audio 6 Audio Interface > Reaper

This means I have the following places where levels need to be set:
Mixer channel trim, Compressor settings, Mixer output, Audio Interface Input.

In general, after recording the podcast, I use a "loudness maximizer" to squash the dynamics so that the levels are all pretty even whether we are talking quietly or being boisterous so that the listeners don't have to keep adjusting their volume. I was hoping that adding the Comp/limiter/exp would reduce how much of that need be done in post and would also make sure to limit the input signal so that, no matter what, there would be no clipping at the mixer.

I'm having a hard time figuring out how to balance setting the levels at all these stages. Like how much trim at the mixer to make sure the compressor is getting enough signal. But then even though the comp makes sure the mixer isn't clipping, it can still end up clipping in reaper if I have the mixer output or the audio interface input set to high.

What would be the best process for getting things optimized here? Thanks!

[nofish]

I'd probably record totally dry and do the processing in Reaper, because if you're (like me) not nearly perfect in setting up compressor/limiter/expander on the mixer once it's (wrongly) done, there's no way to change it anymore. (You could then save presets/FX chains in Reaper to reduce work for the further podcasts.)

About headroom, if you're recording in 24bit you have lots of dynamic range available (over 100db according to here:
http://www.soundonsound.com/sos/sep1...s/qa0910-1.htm)
so not clipping shouldn't actually be a problem.

[Nos402]

Well luckily for the podcast I just want it to limit peaks and squash the hell out of the dynamic range.