Vocal Volume Automation: The Last 5% That Sounds Like a Record
Vocal mixing · 4 min read
Vocal automation comes last, after the full chain, on the track fader. Work in 0.25–1 dB moves drawn across whole phrases, not words. Then automate the reverb and delay sends as well: pulling the send down 3 dB while the fader goes up 1 dB is what makes a line step forward instead of just getting louder.
Move size
0.5 dB
Per phrase · after the chain · sends too
The 60-second answer
Finish the chain first. Automate the post-insert fader in 0.5 dB steps, one shape per phrase, listening at conversation level around 60–70 dB SPL. When a line needs to feel closer, raise the fader 1 dB and drop the reverb send 3 dB. When it needs to feel big, leave the fader and raise the send 4 dB.
Why automation goes after the chain, not before
Clip gain sits before the inserts, so a 2 dB move there changes how hard the compressor works and how much saturation you get. Automation sits after the inserts, so a 2 dB move changes only the balance against the beat and leaves tone and compression where you set them. That means you can ride a chorus up 1.5 dB without the vocal getting thicker, brighter or more compressed. Mix the chain first, then balance it.
Half a decibel is not too small to matter
A 0.5 dB change is roughly the smallest level move most people can hear in a full mix, and it is exactly the size of the move that fixes intelligibility. The word at the end of a phrase that always gets swallowed needs 1 dB, not 3. The line that steps on the snare needs 0.5 dB down. Big moves are for sections, small moves are for words. If your lane looks like a staircase with 3 dB steps, you are writing an arrangement, not mixing one.
How to automate a vocal, step by step
| # | Do this |
|---|---|
| 01 |
Lock the chain before you start EQ, compression, de-essing and effects all set and level-matched. Automation on top of an unfinished chain gets thrown away. |
| 02 |
Turn the monitors down Balance problems hide at loud volume and jump out at quiet volume. Write your moves at conversation level. 60–70 dB SPL |
| 03 |
Write section-level moves first Verse to chorus is usually 1–2 dB. Get the sections sitting right before you look at any single line. Section moves: 1–2 dB |
| 04 |
Then phrase moves One shape per phrase, two nodes rather than eight. Keep the curves gentle so nothing steps. Phrase moves: 0.25–1 dB |
| 05 |
Fix the swallowed words Note every word you cannot make out. Each gets a 1 dB bump over 200–400 ms. Highest-value pass in the whole process. Word bumps: 1 dB |
| 06 |
Automate the reverb and delay sends Down 2–3 dB in dense sections, up 3–6 dB on the last word of a phrase and on held notes. Then check on a phone speaker. Send moves: 2–6 dB |
The trick: automate the send, not just the fader
When you push the vocal fader up 2 dB with a post-fader reverb send, the reverb goes up 2 dB with it. The vocal gets louder and its room gets louder by the same amount, so the ratio of dry to wet never changes — and that ratio is what your ear uses to judge distance. That is why fader-only automation makes a vocal louder but never closer, and why a mix can feel flat even when every level is right. Do it in two moves instead. For an intimate line, raise the fader 1 dB and drop the reverb send 3 dB: the dry-to-wet ratio swings, the vocal snaps forward, and it barely got louder. For the last line of a chorus, leave the fader alone and push the send up 4 dB: the vocal stays where it was and the space around it opens. Two lanes instead of one, and you are automating perspective rather than volume.
Stop guessing the chain.
The Vocal Chain Bible is 88 real vocal chains from Grammy-winning engineers — every plugin, every setting, in order, with free alternatives for each one. 5+ years, 100+ interviews, one PDF.
Get the Vocal Chain Bible — $48 →Mistakes that kill it
- Automating before the chain is finished. Change the compressor threshold later and every move you wrote is wrong by a different amount in every bar.
- Too many nodes. Eight breakpoints inside one word makes the vocal wobble. Two nodes and a curve per phrase is enough.
- Using automation to fix a recording problem. If the singer turned off the mic, the tone changed too. That is clip gain plus a small EQ move.
- Mixing automation loud. At 85 dB SPL everything sounds balanced. Write at conversation level and check loud, not the reverse.
Automation is slow, unglamorous and invisible when it is done well. It is also why a professional mix stays intelligible from a car to a laptop without anyone touching the volume.
Keep going
FAQ
Should I automate the fader or use a compressor?
Both, for different jobs. The compressor handles level changes inside a word. Automation handles changes across phrases and sections, which no compressor can judge musically.
How much automation does a vocal need?
On a three-minute song, expect 30–60 moves: a few section changes, one shape per phrase, and a 1 dB bump on every word you could not make out.
Do I automate before or after the reverb send?
Automate the post-insert fader and the send level separately. Moving them independently is what gives you control over depth as well as loudness.
Who's behind this
Ivan Meshcheriakov · IvanFromRnD
5+ years ago I moved to LA with zero connections and started building from nothing — 14-hour days, meeting people and interning at studios. It took me 5+ years to meet and interview 100+ Grammy-winning engineers — then I turned it all into the Vocal Chain Bible.
Read my full story →