The Ultimate Guide to Mixing Music

Everything you need to know to turn a rough session into a clean, balanced, release-ready mix. Frequency balance, loudness, dynamics, EQ, compression, stereo width, phase, masking, room acoustics, monitoring, and reference tracks — all in one place.

1. What mixing actually is

Mixing is the process of taking multiple recorded tracks — vocals, drums, bass, guitars, synths, whatever — and combining them into a single stereo file where every element is audible, balanced, and emotionally effective. That's the whole job. Everything else — EQ, compression, reverb, panning, automation — is a means to that end.

A good mix doesn't sound "processed." It sounds like the song always existed that way. The vocal sits on top of the instruments without fighting them. The bass and kick drum share the low end without canceling each other. The stereo field feels wide but doesn't fall apart in mono. The quiet parts are quiet enough to mean something, and the loud parts hit hard without clipping. Mixing is the craft of making all of that happen at once.

This guide walks through every major dimension of a mix — from the foundational stuff (gain staging, frequency balance) to the technical stuff (LUFS, true peak, phase) to the environmental stuff (room acoustics, monitoring). Each section links to a deeper article on the MixDiagnose blog, and you can run your own mix through MixDiagnose at any point to see where you stand. Let's start at the beginning.

2. Gain staging: the foundation

Before you touch an EQ or a compressor, you need to get your levels right. Gain staging is the practice of managing the volume of every signal at every point in your signal chain — from the input of a channel strip to the input of a plugin to the output of your master bus. If your levels are wrong, nothing else will work properly. A signal that's too hot will clip your plugins' inputs in ways you can't hear until it's too late. A signal that's too quiet will force you to crank the master, raising the noise floor and reducing resolution.

The practical rule: aim for your individual tracks to peak around −6 to −12 dBFS, and your master bus to peak no higher than −3 dBFS before mastering. This gives every plugin enough headroom to work with and leaves space for the mastering engineer (or the mastering plugin) to do their job. If you're clipping your master bus, you're already in trouble — and it's the single most common reason home mixes sound bad.

→ Read more: Gain Staging: The #1 Reason Your Mix Sounds Bad (And How to Fix It)

3. Frequency balance

Frequency balance is the distribution of energy across the audible spectrum — roughly 20 Hz to 20 kHz. A well-balanced mix has appropriate energy in every region: enough low end to feel weight, enough low-mids for warmth, enough midrange for presence, enough high-mids for clarity, and enough highs for air. When one region is too loud or too quiet, the mix sounds wrong in ways that are sometimes hard to name but easy to feel. A mix with too much 200–500 Hz sounds muddy. A mix with too much 3–5 kHz sounds harsh. A mix with no energy above 10 kHz sounds dull and claustrophobic.

The hardest part of frequency balance is that you can't fix it by ear alone — at least not reliably. Your monitoring environment colors what you hear, and ear fatigue makes you progressively worse at judging balance over a long session. This is where a spectrum analyzer or an automated mix analysis tool becomes essential. MixDiagnose shows you the frequency profile of your mix compared to a target curve derived from thousands of professionally mixed tracks, and flags any region that's more than a few dB off. That gives you an objective starting point: "cut 250 Hz by 3 dB on the bass and the kick" instead of guessing.

A common mistake is trying to fix frequency balance on the master bus. If your low mids are building up, the fix is usually in the individual tracks — the bass, the guitar, the low tom — not a broad EQ cut on the whole mix. Fix it at the source, then verify the combined balance. MixDiagnose shows you the combined balance so you can confirm your fixes worked.

→ Read more: How to Fix a Muddy Mix — Frequency Balance Guide

4. LUFS & loudness

LUFS (Loudness Units Full Scale) is the standard metric for perceived loudness, and it's what every major streaming platform uses to normalize your track. Spotify normalizes to −14 LUFS. Apple Music to −16 LUFS. YouTube to −14 LUFS. If your mix is louder than the target, the platform turns it down — and if you pushed it there with a limiter, you've thrown away dynamic range for nothing. If your mix is quieter than the target, the platform turns it up, which can reveal flaws you didn't hear in the studio.

For mixing — not mastering — the LUFS number matters as a sanity check. A mix that's already at −8 LUFS before mastering is almost certainly over-compressed; there's no room left for the mastering stage to do anything. A healthy mix for mastering typically sits around −18 to −14 LUFS integrated, with peaks no higher than −3 dBTP (true peak). That gives the mastering process space to add loudness without crushing what's left of your dynamics.

True peak (dBTP) is a related but distinct measurement. Your sample-peak meter might say you're at −0.1 dBFS, but the actual analog waveform — reconstructed between samples — can exceed 0 dBFS. This is called an inter-sample peak, and it causes clipping on some DACs and during lossy encoding (MP3, AAC). For streaming delivery, you want true peak no higher than −1 dBTP. MixDiagnose reports both integrated LUFS and true peak so you know exactly where you stand.

→ Read more: What Is LUFS? Loudness Units Explained · Spotify Loudness Specs — Complete Guide · True Peak Explained: Why Your Mix Is Clipping

5. Dynamic range

Dynamic range is the difference between the loudest and quietest parts of your mix. It's what makes music feel alive — the verse is softer than the chorus, the snare hits harder than the ambient pad, the vocal breath has space around it. When you compress too aggressively (or limit too hard in mastering), you squeeze that range down and everything sounds the same loudness. The track feels flat, fatiguing, and small, even though the meter says it's loud.

A useful metric is the crest factor — the difference between peak and RMS (or LUFS) level. A mix with a crest factor of 12–18 dB has good dynamic range. Below 8 dB, you're in over-compressed territory. MixDiagnose reports this directly: if your dynamics score is in the red, you've squeezed too hard somewhere — usually on the master bus compressor or a limiter you added "just to hear how it sounds loud."

The fix is usually to back off. Remove the master bus limiter while mixing. Compress individual tracks to control their dynamics, not to make them louder. Let the mastering stage handle the final loudness — that's its job. A mix with good dynamic range gives the mastering engineer (or tool) something to work with. A mix that's already flat gives them nothing.

6. Compression

Compression is the tool you use to control dynamic range on individual tracks. It reduces the volume of loud peaks so the overall track can be turned up, making the quieter parts more audible. The four main controls — threshold, ratio, attack, and release — determine how and when the compressor reacts. Threshold sets the level above which compression kicks in. Ratio sets how much the signal is reduced (a 4:1 ratio means anything above the threshold is reduced to one-quarter of its excess). Attack sets how fast the compressor responds to a peak. Release sets how fast it lets go.

The most common mistake is using compression to make things louder rather than to control dynamics. If you're dropping the threshold and cranking the makeup gain, you're not compressing — you're just turning up the noise floor and squashing transients. The right approach: set your threshold so the compressor is only catching the peaks you want to tame, choose a ratio that's appropriate (2:1 to 4:1 for most mixing tasks, higher only for effect), and set attack and release to match the instrument. Fast attack on a vocal tames plosives; slow attack on a drum lets the transient through before clamping down.

→ Read more: Compression Basics for Music Producers: When and How to Use It

7. EQ techniques

EQ is the most powerful tool in your mixing toolkit. It lets you shape the frequency content of each track — cutting frequencies that are unwanted or clashing, and boosting frequencies that need emphasis. The guiding principle of mixing EQ is "cut before you boost." Cutting a problem frequency (say, 300 Hz on a bass guitar that's muddying the mix) almost always sounds more natural than boosting the opposite frequency. Cuts are surgical; boosts are obvious.

High-pass filtering is your first move on almost every track that isn't a bass instrument. Vocals, guitars, keyboards, even snare drums often have unnecessary sub-bass content below 80–100 Hz that does nothing but eat headroom and muddy the low end. A gentle high-pass filter at 80 Hz on a vocal, 100 Hz on a guitar, and 40 Hz on a kick drum clears up the low end dramatically without affecting the character of the instruments.

The frequency regions to know: 20–60 Hz is sub-bass (kick drum fundamentals, bass sub). 60–200 Hz is bass (body of bass guitar, kick beater). 200–500 Hz is the mud zone — build-up here makes everything sound boxy. 500 Hz–2 kHz is the midrange (vocals, guitars, snare body). 2–6 kHz is presence (vocal intelligibility, attack). 6–12 kHz is sibilance and harshness. 12–20 kHz is air and sparkle. MixDiagnose flags any of these regions that are significantly out of balance.

→ Read more: How to Use EQ to Fix Your Mix: A Producer's Guide

8. Stereo width

Stereo width is how far your mix spreads across the left-right field. A narrow mix (everything panned center) sounds cramped and small. A wide mix (elements panned hard left and right) sounds big and immersive. But width is a trade-off: the wider you go, the more you risk phase problems and mono incompatibility — which is a serious issue when most casual listening happens on phone speakers and Bluetooth speakers that sum to mono.

The conventional approach: keep your core elements — kick, bass, lead vocal, snare — in the center. Pan supporting elements (guitars, keys, background vocals, percussion) to the sides. Use stereo widening plugins or Mid/Side processing on groups of instruments to push them wider, but always check the result in mono. If the element disappears or sounds thin when you hit the mono switch, you've gone too wide. MixDiagnose measures your stereo width across the frequency spectrum and flags regions where the image is too narrow or too wide for the genre.

→ Read more: Stereo Width in Mixing: How Wide Is Too Wide?

9. Phase issues & mono compatibility

Phase issues happen when two versions of the same signal are time-offset from each other, causing their waveforms to partially cancel. The most common sources: multi-mic drum recordings (the overhead mics and close mics are at different distances), DI and amp signals on bass, and stereo widening plugins that use delay-based widening. When the signals are in phase, they add constructively (louder). When they're out of phase, they cancel (quieter, thinner, hollow).

The test is mono. Sum your mix to mono and listen. If elements get quieter, thinner, or disappear entirely, you have phase problems. The fix is usually time-alignment: nudge the later track earlier until the waveforms line up, or use a phase alignment tool. On stereo widening, pull back the width until the mono sum sounds full again. A correlation meter reading below 0 means your left and right channels are actively canceling — that's a red flag.

→ Read more: Mono Compatibility: Why Your Mix Disappears on Phone Speakers

10. Frequency masking

Masking is when one instrument's frequency content covers up another's, making the masked instrument harder to hear even though it's "there" in the mix. The classic example: bass guitar and kick drum both have energy at 60–100 Hz. If they're both loud in that range, they fight, and neither sounds clear. The fix is frequency carving — give each instrument its own space. The kick gets 60 Hz (the fundamental thump); the bass gets 80–100 Hz (the body). Cut the bass slightly at 60 Hz where the kick lives, and cut the kick slightly at 80–100 Hz where the bass lives.

Masking happens across the spectrum, not just the low end. Guitars and vocals mask each other in the 1–3 kHz range. Multiple synths mask each other everywhere they overlap. The general strategy: identify the most important frequency of each instrument (the one that defines its character) and protect it by cutting that frequency on the competing instruments. This is easier said than done, which is why automated analysis helps — MixDiagnose identifies regions where multiple instruments are likely fighting and suggests specific cuts.

11. Room acoustics

Your room is part of your mix. Every surface reflects sound, and those reflections combine with the direct sound from your monitors to create the frequency response you actually hear. An untreated room typically has massive peaks and nulls in the low end — sometimes 20 dB swings — meaning you're mixing in a distorted version of your own sound. If your room has a 10 dB peak at 80 Hz, you'll mix the bass too quiet to compensate, and the mix will sound thin everywhere else.

The fixes, in order of impact: bass traps in the corners (this is the single biggest improvement you can make), absorption at the first reflection points on the side walls and ceiling, and a symmetric setup so the left and right sides of the room match. You don't need a fully treated studio — even a few bass traps and some first-reflection panels will dramatically improve what you hear. Room correction software (Sonarworks, Dirac) can measure and compensate for the remaining issues, but it works best on a room that's already been physically treated.

If you can't treat your room, mix at low volumes (where room reflections matter less), use headphones for a second perspective, and rely on reference tracks to calibrate your brain. And run your mix through MixDiagnose — it doesn't care about your room, so it gives you an objective check that your ears can't.

12. Monitoring & listening levels

Monitoring is the chain of equipment and practices you use to listen to your mix: your speakers (monitors), your headphone setup, your listening level, and your habit of checking on multiple systems. The goal isn't a perfect speaker — it's a known speaker. You need to know what your monitors sound like so well that you can predict how your mix will translate. That comes from listening to reference tracks on them constantly.

Listening level matters more than people think. The Fletcher-Munson curve describes how human hearing perception changes with volume — at low levels, bass and treble seem quieter relative to mids. At high levels (above 85 dB SPL), everything sounds flattered and exciting. The sweet spot for mixing is around 75–80 dB SPL — loud enough to hear the full spectrum accurately, quiet enough to avoid ear fatigue and the false excitement of high volume. Mix at a consistent level. Check at low volume to verify the vocal and the hook are still audible. Check loud only briefly to confirm the low end hits.

Always check your mix on at least three systems: your main monitors, headphones, and a consumer device (phone speaker, earbuds, car stereo). If the vocal disappears on earbuds, you have a midrange problem. If the bass vanishes on the phone, your low end is too sub-dependent. If it sounds great everywhere, you're done.

13. Reference tracks

A reference track is a professionally mixed and mastered song in the same genre as yours that you use as a sonic benchmark. It's the single most underrated tool in mixing. When you've been in a session for hours, your ears adapt — you lose perspective on what "good" sounds like. A reference track snaps you back to reality. You A/B between your mix and the reference: is your bass as controlled? Is your vocal as present? Is your stereo field as wide without falling apart in mono?

The technical way to use a reference is to level-match it to your mix (so you're comparing balance, not loudness) and analyze both with the same tool. MixDiagnose does this automatically — it compares your frequency profile, loudness, dynamics, and stereo width against genre-appropriate reference curves derived from professional tracks, and tells you exactly where you differ. That turns the vague feeling of "my mix doesn't sound as good as the reference" into specific, actionable fixes: "your low mids are 4 dB hotter than the reference — cut 250 Hz."

Choose references in the same genre, with similar instrumentation, at a similar target loudness. Don't compare your indie folk mix to a Calvin Harris track. The reference is a target, not a miracle — it shows you the gap, and your job is to close it.

14. Pre-mastering checklist

Deep-dive articles