A vocal that sits well in a mix is almost never one miracle processor: it’s a sequence of small decisions chained in the right order. We call it a vocal chain. The idea is simple: each link prepares the next, and the order matters as much as the settings. That’s the difference between a voice that stays in front without tiring the ear and one you keep pushing too hard or not enough. Here’s a clear, link-by-link method to build it without losing your evening to it.
Start from a clean take
No processing chain fixes a bad capture. The foundation is set at the source: a well-placed microphone and a preamp at the right gain save you from compensating later. In the mix, the first move isn’t to add but to clean up: roll off useless sub energy with a high-pass filter (often 80-120 Hz on a voice), find and tame distracting breaths or mouth noises, and check your gain structure so you hit the next stages at a comfortable level.
The order of the links
There’s no single order carved in stone, but one arrangement is widely agreed because it works in the vast majority of cases. Here’s the logic.
- Subtractive EQ: before compressing, remove what gets in the way — a nasal resonance, a boxy bump in the low mids. Compressing a cluttered voice asks the compressor to react to faults rather than to the voice itself.
- Compression: this is the heart of the chain. It narrows the gap between loud and soft syllables so the voice holds its place. On vocals, a moderate ratio and a not-too-fast attack keep things natural; for the settings themselves, see our guide to setting a compressor. Many engineers split the work across two gentle compressors rather than one that crushes.
- De-essing: compression often wakes up sibilance. A de-esser placed after the compressor tames the “s” and “sh” sounds without dulling the top end.
- Corrective and tone EQ: once the dynamics are under control, you can open the top for air, add a touch of presence around 3-5 kHz, or dip a region that masks another element.
- Gentle saturation: a hint of harmonics adds density and helps the voice cut through on small speakers, without raising the level.
Effects on sends, not inserts
Reverb and delay are almost always handled on aux sends, on separate effect tracks, rather than as inserts on the voice. The benefit is twofold: you balance the effect independently of the dry level, and you can treat the effect return (a high-pass on the reverb, a filtered delay) so it doesn’t blur intelligibility. The type of reverb matters as much as the amount: a short plate glues the voice to the mix, a long hall stages it — our piece on reverb in the mix unpacks those trade-offs.
Automation, the link people forget
Compression handles fast swings; automation handles swings of intent. A swallowed word, a chorus that must ride over a denser guitar, a phrase that trails off at the end: nothing replaces a few hand-drawn volume automation points. That’s often what separates a “processed” voice from one that’s genuinely mixed.
Common pitfalls
Three mistakes come up again and again. First: stacking processors while ignoring each one’s output gain, so you judge “better” what is simply “louder”. Always compare at matched level. Second: over-compressing, then turning the level back up, which brings breath and mouth noise to the front. Third: dialling the voice in solo. A voice is set in the mix, among the other elements, because that’s where it will live. The same discipline of order and restraint shows up later in the chain, at the mastering stage.
My take
The best vocal chain is often the shortest one that does the job. It’s tempting to add one more link out of doubt; most of the time that doubt belongs to subtractive EQ or to the take, not to an eighth plug-in. Keep in mind that order structures everything: clean up, control the dynamics, then colour. If you keep one reflex, make it comparing each stage at matched level — it’s the simplest move, and the one that improves a mix vocal fastest.
