Before I touch a fader, I spend about twenty minutes on the session itself: route every track to a group bus and the groups to the mix bus, gain stage so each channel hits its first plugin around minus 18 dBFS on average, build VCA faders for the sections, color and order the tracks the same way every time, and set one fixed monitoring level to judge loudness against. A saved template turns that into five minutes. The records mixed this way are at /credits; the setup below is what stops a mix from stalling on its own housekeeping.
Why gain stage before anything else?
Because analog-modelled plugins expect a particular level, and a hot channel pushes them past it. Many compressor and console emulations are calibrated so that minus 18 dBFS at the input lines up with the 0 VU sweet spot they were modelled on, and feeding them a signal that peaks near 0 dBFS drives them into a distortion you did not ask for (gain staging). I set clip gain or a trim plugin first in each chain so the track averages around minus 18 dBFS and peaks near minus 10, then leave the channel fader at unity. Sound On Sound frames the same idea across the whole path, from mic pre to mix bus (Sound On Sound on gain staging), and Sonarworks lays out the tracking-to-mix version (gain staging guide). What I want at the end is a mix bus reading around minus 6 dBFS with every fader at zero, before I have made a single balance decision.
How should the routing be structured?
Every source goes to a group bus, and the groups go to the mix bus. Drums to a drum bus, all the vocals to a vocal bus, the music to an instrument bus, effects returns to their own bus. That leaves me three or four faders that move whole sections, a place to put glue compression on the drum and vocal groups, and one mix bus at the end. Group compression on the vocal bus follows the same low-ratio approach as parallel compression on a vocal, a couple of dB of gain reduction to hold the section together rather than to reshape any single track. When a reverb is shared, it lives on its own bus and every source sends to it, so one control sets the depth of the whole mix.
What do VCA faders do that a group bus does not?
A VCA moves the level of a set of channels without collapsing their internal balance or their sends. If the second chorus needs to come up 2 dB, one VCA fader lifts the whole band while the reverb and delay sends track the change with it, which an audio group fed post-send does not always do. I build a VCA per section, one each for drums, bass, music, lead and backing vocals, and one master VCA over all of them. That master VCA is where a fade at the end of a song lives, so the automation sits in one place instead of on twenty tracks.
How do you order and color the tracks?
The same way in every session, so my hands already know where the snare is. My order runs top to bottom: drums, percussion, bass, guitars, keys, lead vocal, backing vocals, effects. Each group gets one color, so a glance tells me which section I am scrolled into. This is the same track order I use when I prepare the folder that leaves for a mix engineer, a separate step that happens after the mix. Consistent order saves real time: when every session looks alike, the muscle memory carries between songs and the mix starts on the music instead of on a search.
What level should you mix at?
One fixed monitor level, chosen once and marked on the controller. I set the room so a full mix around minus 6 dBFS sits at a comfortable, repeatable loudness, then judge every balance against that same setting (reference monitoring level). Loudness perception shifts with level, and the low end and the top seem to rise as you turn up, so a balance decided at a random volume comes apart on other systems. I keep a second, quieter level for checking the vocal, because a lead that reads clearly at a low monitor level is usually sitting at the right height in the mix.
What gets saved into the template?
The buses, the VCAs, the colors, the track order, the monitor level, and a set of empty channels already routed. I keep a couple of default inserts loaded and bypassed on the group buses, a bus compressor on the drums and a gentle one on the mix bus, so they are there the moment I want them. The rough mix and the reference imports come in per song rather than living in the template; that method is in reference tracks that work. Opening the template and importing a prepared session means the first sound I hear is already gain staged and routed, and the twenty-minute setup has become five.
Frequently asked
What level should tracks peak at going into a mix? Aim for an average near minus 18 dBFS and peaks around minus 10, with the channel fader at unity. That keeps analog-modelled plugins in the range they were calibrated for and leaves headroom on the mix bus.
Do I need VCA faders if I already have group buses? They do different jobs. A group bus sums audio and is where you compress a section. A VCA controls the faders of a set of channels and their post-fader sends together, which is better for riding a whole section without changing its internal balance.
Should I set my monitor level once and leave it? Yes for the main mix level, with one quieter setting for checking the vocal. A fixed reference makes your loudness judgments repeatable, so a balance holds up when it leaves the room.
If your mixes start slow
Send me the session and one line on where it stalls. What comes back is a routing and gain-staging template set up the way above, with the buses and VCAs already built, so the next song starts on the music. Background and contact at /about.