First I apply compressors to both audio channels (usually it's just mine and Geoff's) and limit the master output to 0 dB. At first I just throw these modules in and don't tweak the settings just yet.
This is the time-consuming part. I basically go through the audio with a fine-toothed comb. This consists of listening to the entire podcast, trimming out silence and "uh" and "mm" and all that mumbly crap.
I also cut out parts where we messed up or if something was just not remotely interesting. Geoff and I are both really rambly, so sometimes I cut out entire conversations!
Basically my goal with this step is to get the podcast sounding lean and compact with almost no silence.
Next I'll go back and tweak the compressors. I'm not an audio engineer, just a hobbyist, so I'm going for a "good enough" approach here. I want to ensure that everyone on the podcast sounds loud enough to be heard clearly at normal volumes, at all times.
I don't do this step for every podcast, but sometimes I like to go in and add some additional polish in the form of sound effects. Sometimes I'll add reverb to someone's voice just for fun (like when an announcement is made), or I'll add some sound effects from Onslaught! Arena between conversation topics to give the listener a quick breather.
The last step is to include the music. I like to start the podcast off with a song, but fade it out quickly. I used to include quiet music behind the whole 'cast, but people seemed to hate that!
Lastly I really enjoy playing the show out with a song from Joshua Morse because I think he's an amazing artist and want to spread his music around.