Mixing is the stage where a rough demo starts sounding like a record. It’s also the stage most of us learn backwards — collecting random EQ tricks from YouTube before understanding why the mix feels wrong in the first place.
I rewrote this guide in 2026 to cut the fluff. No magic plugins, no secret chains. Just the order I work in on every track, with actual numbers you can start from and adjust by ear.
What mixing actually is
Mixing means balancing every track in your session — levels, panning, EQ, compression, effects — so each part is heard clearly and the whole thing works on any speaker. That’s it. It’s not sound design, and it’s not mastering.
Before you touch anything, know the role of every track. Is that vocal the lead or a background stack? Is that synth the hook or just texture? A part that only exists as texture shouldn’t fight for the same space as the hook. Most muddy mixes come from tracks that were never assigned a clear job.
One more honest point: about 80% of a good mix is just levels and panning. If your mix falls apart when you bypass every plugin, the problem isn’t your EQ settings — it’s the balance underneath.
Prepare the session before you touch a fader
Ten minutes of prep saves hours later. Here’s my checklist:
- Clean the tracks. Cut clicks, pops, hum and dead air from recordings. Fade the edges of every clip (5–10 ms) so nothing clicks at the boundaries.
- Name and color-code everything. “Audio 27” tells you nothing at 2 a.m. Group tracks into buses: drums, bass, music, vocals, FX. Four or five buses is enough for almost any track.
- Gain-stage. Pull faders down so each track averages around -18 dBFS and your master bus peaks below -6 dBFS before any processing. You want headroom, not a master already touching zero.
- Pick 2–3 reference tracks in the same genre, and check against them at matched loudness throughout the mix. Your ears drift; references don’t.
- Freeze or bounce heavy synths. A session that stutters is a session you can’t judge. If Serum instances are eating your machine, my Serum 2 CPU optimization guide covers the multicore and quality settings that matter.
And the biggest time-saver of all: start with better sounds. A patch that already has its EQ and dynamics designed in needs almost no surgery later. That’s honestly the reason I make presets for a living. If you’re on Serum 2, grab the free Serum 2 preset pack — 250+ sounds — and notice how much less corrective EQ they need compared to raw init patches. For a wider palette across genres there are also our Serum 2 preset bundles.
A step-by-step mixing order
Everyone develops their own workflow eventually. This is the order that keeps me out of trouble:
1. Drums first
Drums are the skeleton, so build them before anything else. Set the kick as your anchor — I aim its peaks around -8 dBFS — then balance the snare against it, then hats and percussion around those two.
Useful frequency landmarks: kick fundamental lives at 50–60 Hz and its click at 3–4 kHz; snare body sits near 200 Hz and the crack near 2 kHz. High-pass overheads and hats around 300–400 Hz so they stop fighting the kick. For glue, a bus compressor at 4:1, attack 10–30 ms, fast release, doing 2–4 dB of gain reduction is a safe start.
Also: no EQ fixes a weak sample. If your drums sound small at the source, swap the source. I keep processed one-shots and loops around for exactly this — the Drums Utopia kit is what I reach for on house and future bass sessions.
2. Bass and 808s
The kick–bass relationship makes or breaks the low end. Two rules that solve most problems: keep everything below 100–120 Hz in mono, and never let the kick and bass occupy the same frequency at the same moment.
Practically, that means either sidechain the bass to the kick (2–3 dB of ducking is plenty — it should be felt, not heard) or carve a few dB out of the bass right where the kick’s fundamental sits. High-pass every non-bass track at 100 Hz or higher; the low end clears up instantly.
For 808s specifically, saturation beats boosting. Drive adds harmonics at 100–300 Hz that let the 808 read on phone speakers. I wrote a full walkthrough in the Serum 2 808 bass guide, and if you’d rather skip the design stage, The best 808 Vol.1 is a kit of tuned, pre-saturated one-shots.
3. Music: synths, keys, guitars
With the rhythm section solid, bring in the melodic layers. High-pass aggressively — most pads and guitars lose nothing below 150–200 Hz. Pan for contrast: double guitars hard left and right, keep the lead synth centered, spread pads wide. If two parts mask each other, don’t reach for EQ first — try re-panning, octave-shifting one part, or muting one of them. Arrangement fixes beat mixing fixes every time.
4. Vocals
Level automation comes before compression. Ride the vocal clip by clip so it sits evenly, then compress gently — 3–6 dB of gain reduction at 3:1 or 4:1. De-ess after compression, since compressors push sibilance forward. A short slap delay (80–120 ms) usually keeps a vocal more upfront than reverb does.
5. Effects
Put reverb and delay on send/return tracks, not inserts, so multiple tracks share the same space. Then filter the returns: high-pass around 300 Hz, low-pass around 8 kHz. Filtered reverb gives you depth without the wash. If you’re building effects inside the synth instead, Serum 2’s new FX bus system is genuinely powerful — I covered routing and the standalone SerumFX version in the Serum 2 effects guide.
6. Balance passes and automation
Now loop the track and make small moves: dip the guitars 1 dB under the vocal, ride the pad up in the bridge, open a filter into the drop. Automation is what makes a mix feel alive. Check against your references at matched loudness, take a break, and listen again the next day on different speakers before you call it done.
Starting points that actually work
These aren’t rules — they’re defaults to adjust by ear. But they’ll get you in the neighborhood fast:
| Element | Move | Why |
|---|---|---|
| Kick | Boost 50–60 Hz and 3–4 kHz, cut 300–500 Hz | Weight plus click, less cardboard |
| 808 / bass | Saturation before EQ; mono below 120 Hz | Audible on small speakers, stable low end |
| Snare | Boost ~200 Hz (body) or ~2 kHz (crack) | Pick one — boosting both gets harsh |
| Vocals | HPF 80–100 Hz, cut 300–500 Hz if boxy, shelf +2 dB above 10 kHz | Clarity and air without sibilance |
| Pads / guitars | HPF 150–200 Hz, pan wide | Space for kick, bass and vocal |
| Reverb returns | HPF 300 Hz, LPF 8 kHz | Depth without mud or hiss |
Common mixing mistakes
- Too many layers. Eight pads stacked together sound smaller than two good ones. If a track doesn’t earn its place when soloed against the mix, mute it. Density is not the same as fullness.
- Ignoring gain staging. If everything is slammed into the red before the mix starts, every plugin afterwards behaves worse. Leave headroom; you can always turn the master up at the end.
- EQ boosts everywhere. Cut more than you boost. A 2 dB cut in the right place beats a 6 dB boost in the wrong one, and ten boosted tracks just make a louder, harsher version of the same mud.
- Over-compression. If the meters barely move and the track feels flat, back off. Most sources need 2–4 dB of gain reduction, not 10.
- Fake width. Stereo wideners on everything collapse to nothing in mono. Get width from panning, doubles and short delays — then check the mix in mono. If the hook disappears, fix it now, because club systems and phone speakers will find that problem for you.
- No references. Your room and your ears lie, especially after two hours. A reference track at matched loudness is the cheapest mixing tool that exists, and almost nobody uses it consistently.
Mastering, briefly
Mastering is the final polish on the finished stereo mix — not a rescue mission. If the mix is wrong, go back and fix the mix.
A simple 2026 chain that works: gentle bus EQ (moves of 1 dB or less), a glue compressor doing 1–2 dB of reduction, optional saturation or a limiter-side clipper, then a limiter with the true-peak ceiling at -1 dBTP. Loudness targets depend on the destination: around -14 LUFS integrated matches streaming normalization, while club-oriented genres are commonly mastered louder, around -8 to -10 LUFS. Louder always costs dynamics, so push until it hurts the punch, then back off.
Export the mix you send to mastering as a 24-bit WAV with 3–6 dB of headroom and no limiter on the master bus.
FAQ: mixing music
How long does it take to mix a song?
For a producer mixing their own track, anywhere from two hours to two days. Speed comes from good source sounds and a repeatable order of operations, not from working faster. If a mix isn’t coming together after a few hours, sleep on it — fresh ears solve more problems than new plugins.
Can I mix on headphones?
Yes, and in an untreated room headphones are often the more honest option. The catch is stereo width and low end read differently on headphones, so cross-check on at least one other system — car speakers and a phone speaker are a surprisingly useful pair. Mixes that survive all three are done.
What level should I mix at?
Keep individual tracks averaging around -18 dBFS and your pre-master mix peaking below -6 dBFS. Don’t chase a LUFS number while mixing — loudness is the mastering stage’s job. Monitor at a moderate volume (roughly conversation level) and only check loud briefly; your ears flatten frequency response at high volume and every mix starts sounding great.
What’s the difference between mixing and mastering?
Mixing balances the individual tracks inside a session. Mastering processes the finished stereo file — final EQ, glue compression and limiting — so it translates across systems and hits a sensible loudness for its destination. They’re separate jobs, and doing them in separate sessions (or on separate days) keeps you objective.
That’s the whole system: prep the session, build from drums up, cut more than you boost, and reference constantly. If you want a shortcut on the source-sound side, the free Serum 2 presets are there — 250+ patches that already sit right in a mix, so you can spend your time on the balance instead of the surgery.
I run Monosounds.studio and make the packs myself. Ten-plus years of producing, 97+ releases, 16,000+ downloads. Every Serum 2 preset we ship passes a CPU check on a mid-range laptop and gets used in a real track before release. If it does not survive a mix, it does not go in the pack.
More about Maxim & the studio →
