Mixing Loudness Standards: Where Should Your Mix Peak

Mixing Loudness Standards: Where Should Your Mix Peak

Your mix should peak below clipping, usually around -6 to -3 dBFS on the stereo bus before mastering, with no true-peak overs, no limiter damage, and enough dynamic range that the mastering stage can work. Streaming loudness numbers like -14 LUFS are playback and mastering references, not a rule that every mix should be printed to. A good mix is clean, balanced, and controlled before it is loud. If the mix already has distortion, crushed drums, or a clipped vocal bus, a perfect LUFS number will not save it.

Loudness advice gets confusing because people use the word "mix" when they mean "master." A mix engineer needs headroom, balance, and clean peaks. A mastering engineer needs a finished mix that can be shaped, leveled, and checked for delivery. Those are related, but they are not the same target. The mix does not have to be loud enough for Spotify. It has to be healthy enough to become loud later.

Spotify's artist guidance explains loudness normalization and mastering tips around integrated LUFS and true peak. That matters for release delivery. But when you are still mixing, the more useful question is simpler: does the stereo bus clip, do the loud sections still punch, does the vocal survive level changes, and does the export leave room for final mastering?

The Short Answer: Mix Peak vs Master Peak

For a normal pre-master mix, leave conservative headroom. A stereo mix peaking around -6 to -3 dBFS is usually easy to master as long as it is not clipped and the balance is right. The exact number is less important than the condition of the file. A mix peaking at -4 dBFS with clean transients is better than a mix peaking at -10 dBFS with clipped tracks inside the session.

For a final master, true peak and loudness targets become more important. Spotify talks about -14 dB integrated LUFS and true-peak ceilings in its mastering tips, but it also notes that louder masters should leave more true-peak margin to avoid extra distortion during encoding. That is mastering guidance. Do not force your unfinished mix into those numbers with a limiter just because a platform page mentions them.

Why True Peak Matters

A normal sample-peak meter only shows sample values. True peak estimates what can happen between samples when the audio is reconstructed during playback. That matters because a file that never shows 0 dBFS on the sample meter can still exceed 0 dBTP after conversion, limiting, or lossy encoding. Youlean's loudness documentation explains true peak as a way to catch peaks between sample points that simple meters can miss.

In a mix session, you do not need to obsess over every tiny true-peak movement if you are leaving healthy headroom. But if your rough mix or reference limiter is pushing near 0, true peak becomes important fast. A mix that appears safe at -0.2 dBFS sample peak can create problems later. A mix peaking at -5 dBFS without clipped internal buses gives mastering room to work.

Use true peak as a warning light, not a creativity limiter. If the meter says your rough bounce crosses 0 dBTP, back off the limiter, clipper, or output gain. If the final pre-master is sitting several dB below clipping and sounds balanced, you are probably fine.

Do Not Chase -14 LUFS While Mixing

The most common loudness mistake is mixing into a final streaming target. A producer reads that Spotify normalizes to -14 LUFS, then pushes the mix bus until the integrated meter hits -14. That can create a false sense of safety. The mix may hit the number while the kick is too loud, the vocal is buried, the snare is clipped, and the chorus collapses under limiting.

LUFS is useful during mixing, but not as a pass/fail target. Use it to compare a rough mix against references at matched volume. Use it to notice whether a hook is much denser than the verse. Use it to see whether the rough limiter is doing too much. Do not use it to decide that the mix is finished.

When you want to hear what the song might feel like after mastering, put a limiter on the monitoring chain or a temporary master bus, print a separate reference bounce, and keep the clean pre-master export separate. That way the artist can hear a loud version without forcing the mastering engineer to work from the crushed one.

A Practical Mix-Bus Peak Workflow

Use this workflow before exporting the mix for mastering:

1. Bypass any limiter that is only there for loudness.
2. Play the loudest chorus and watch the stereo bus peak.
3. If the bus clips, pull down groups or the mix bus gain, not just the final output after the damage.
4. Check the vocal bus, drum bus, instrument bus, and FX bus for hidden clipping.
5. Listen to the kick, snare, vocal, and bass after lowering level to make sure the balance still works.
6. Export a clean pre-master at the session sample rate and bit depth requested.
7. Export a separate loud reference if needed, labeled clearly as reference only.

The important detail is step three. Pulling down the final fader after a plugin has already clipped does not undo the clipping. You need to find where the overload starts. Sometimes the stereo bus is fine, but the drum bus is clipped by a saturator. Sometimes the vocal bus is clean, but a delay return overloads into the chorus. Fix the first clipped stage.

Where Different Genres Usually Land

Genre affects loudness expectations, but it should not override clean mixing. Dense trap, drill, EDM, and modern pop often use louder references. Indie, acoustic, jazz, worship, singer-songwriter, and cinematic music may need more space and dynamic range. The pre-master mix still should not clip in either case.

These are not rules carved in stone. A clean mix peaking at -8 dBFS is usable. A clean mix peaking at -2 dBFS can still be usable if it has no clipping and the mastering engineer approves. The danger is not the number by itself. The danger is hidden distortion, no headroom, and level decisions made by a limiter instead of by the mix.

How Loud Should the Rough Mix Be?

A rough mix for artist approval can be louder than the clean pre-master. That is normal. Artists often compare your bounce against released songs, not against other unmastered mixes. If you send only a quiet pre-master, they may think the mix is weak even when the balance is good.

The safe method is to export two files:

• Clean pre-master: no final limiter, no clipping, healthy headroom, intended for mastering.
• Loud reference: temporary limiter or clipper for vibe checking, clearly labeled as a reference.

Do not let the loud reference become the delivery master unless that is the actual plan. If the reference limiter changes the vocal level, distorts the low end, or makes the chorus smaller, revise the mix before mastering. The rough loudness should reveal problems, not hide them.

What to Check Before Sending the Mix

Before mastering, check the mix like a file, not only like a song. Listen from start to finish. Watch meters. Check the loudest and quietest sections. Make sure the export starts cleanly and ends cleanly. If there are intentional clipped sounds, write them in the notes so they are not mistaken for mistakes.

Use this checklist:

• No red lights on tracks, buses, or the stereo output.
• No accidental limiter on the master export.
• Clean pre-master peaks below clipping with comfortable headroom.
• True peak does not cross 0 dBTP on any reference-limited bounce.
• Vocal is intelligible in the loudest chorus and quietest verse.
• Low end does not disappear when the rough limiter is bypassed.
• Mono check does not collapse the vocal, bass, or snare.
• File is named clearly and exported at the requested sample rate and bit depth.

The pre-mastering checklist is useful if you want a more detailed file-prep pass. If the issue is not file prep but mix balance, fix the mix first. Mastering should not be asked to solve a vocal that is 3 dB too quiet.

How Normalization Changes the Loudness Decision

Streaming normalization means louder files are often turned down during playback. Spotify explains that it uses loudness normalization to balance softer and louder songs, and its artist guidance includes different listener normalization settings. The practical result is that extra limiting does not always make a track feel bigger to the listener. It can make it smaller after the platform turns it down.

A dynamic mix can feel louder after normalization because the drums still punch, the vocal is clear, and the chorus opens up. A crushed mix may read high on a meter but feel flat after level matching. That is why the best loudness standard for mixing is not "hit the loudest possible number." It is "make a balanced mix that can become loud without falling apart."

If you are mastering yourself, use the LUFS normalizer guide and the mastering limiter settings guide after the mix is already healthy. If you are sending the song out, send the clean mix and a loud reference so the mastering engineer understands the intended energy.

Common Loudness Mistakes

Final Mix Peak Rule

Print the clean mix with no clipping, no accidental limiter, and enough headroom that mastering can make decisions. Around -6 to -3 dBFS peak is a practical zone, but do not panic if the clean mix lands a little outside it. The real standard is whether the file is undamaged and the balance survives when you turn it up, turn it down, check mono, and compare against references.

If you are not sure, send both the clean pre-master and the loud reference to whoever is mastering the song. If you are handling both mixing and mastering yourself, separate the steps anyway. Make the mix work first. Then make it loud. That separation prevents most loudness mistakes before they become release problems.

How to Use Reference Tracks Without Copying Their Loudness

Reference tracks are useful, but only when they are level-matched. A released reference has already been mastered. Your clean mix has not. If you compare them at raw playback level, the released song will almost always feel better because it is louder, denser, and finished. That can push you into unnecessary limiting before the mix is ready.

Turn the reference down until it feels similar in loudness to your mix. Then compare relationships: vocal to kick, snare to vocal, bass to kick, reverb to dry signal, chorus energy to verse energy, and top-end brightness. You are not trying to make your unmastered mix hit the same final integrated LUFS. You are trying to understand whether the balance will survive mastering.

Useful reference questions:

• Does my vocal sit too far behind the beat at matched volume?
• Does my kick feel louder than the reference but less punchy?
• Does my chorus feel smaller even before mastering?
• Is my low end using too much headroom for the amount of weight it gives?
• Do I need mix changes, or am I only missing final mastering level?

This is where many loudness decisions become mix decisions. If the low end eats 6 dB of headroom but does not feel bigger, fix the kick and bass relationship. If the vocal disappears when you turn the mix down, fix vocal level and masking. Do not ask a limiter to solve those problems later.

Meter Setup for a Home Studio

A good home setup needs a peak meter, a true-peak meter, a loudness meter, and your ears. You do not need to stare at numbers all day, but you do need a reliable way to catch problems before export.

Set up your meters like this:

• Keep the DAW peak meter visible on the stereo output.
• Use a loudness meter on the monitoring chain or master bus for reference checks.
• Use true-peak metering on any loud reference bounce or final self-master.
• Check integrated loudness only after playing the whole song or full section.
• Check short-term loudness around verses and hooks to see section contrast.

Momentary numbers can jump around too much to guide final decisions. Integrated loudness tells you the average over time, while short-term loudness helps compare sections. True peak tells you whether the loud version has encoding or playback risk. Together, those meters give context. They do not replace listening.

Low End Is the Loudness Trap

Low end can consume headroom without sounding loud. A sub-heavy 808, kick tail, or bass resonance may push the mix bus close to clipping while the song still feels quiet. The mixer then adds more limiting, which makes the drums smaller and the vocal duller. The problem was never final loudness. It was low-end control.

When the mix peaks too high, check the low end first:

• Mute the 808 or bass and watch how much headroom returns.
• High-pass non-bass instruments so they do not stack rumble.
• Shorten kick or 808 decay if tails overlap.
• Use volume automation instead of flattening bass dynamics with extreme compression.
• Check small speakers so bass harmonics still translate.

A controlled low end lets the master get louder with less damage. An uncontrolled low end makes every limiter work harder than it should.

When to Stop Mixing and Start Mastering

Stop mixing when the balance works without the final limiter, the loudest section feels exciting, the quietest section still communicates, and the clean export has no technical problems. Start mastering when the remaining questions are final loudness, tonal translation, sequencing, delivery files, and platform safety.

If you are still changing vocal volume, kick level, arrangement mutes, reverb amount, or bass notes, you are still mixing. If you are checking true peak, final level, fade shape, alternate versions, and file delivery, you are mastering. Keeping those decisions separate makes the entire release cleaner.

What If the Client Wants It Loud Now?

Send a loud reference, but keep the clean mix. This solves the communication problem without damaging the source. The client hears the energy they expect, while the mastering engineer still receives a file with room to work. Label the files clearly: `song-title-clean-pre-master.wav` and `song-title-loud-reference.wav` is better than `final.wav` and `final2.wav`.

Tell the artist what each file is for. The loud reference is for vibe and approval. The clean pre-master is for final mastering. That one sentence prevents the wrong bounce from becoming the release file.

FAQ

Where should my mix peak before mastering?

A clean pre-master mix usually works well when it peaks around -6 to -3 dBFS, with no clipping on tracks, buses, plugins, or the stereo output. The exact number matters less than sending an undamaged file with enough headroom for mastering.

Should I mix to -14 LUFS?

No. Use -14 LUFS as streaming and mastering context, not as a required mix target. During mixing, focus on balance, headroom, punch, and no clipping. Check LUFS for comparison, but do not force the mix into a final loudness number.

Can I leave a limiter on the mix bus?

You can use a limiter for a loud reference bounce, but send a clean pre-master without the final loudness limiter unless the limiter is an intentional part of the approved sound. Label the files clearly so the reference does not replace the mastering source.

What is true peak in mixing?

True peak estimates peaks between digital samples after reconstruction. It matters most when audio is near clipping or being encoded. If your clean mix has several dB of headroom, true peak is usually safe, but loud references and masters should be checked with a true-peak meter.

Is -6 dB headroom required?

No. It is a useful guideline, not a law. A clean mix peaking at -4 dBFS or -8 dBFS can be fine. What matters is no clipping, no hidden bus distortion, clear balance, and enough room for mastering moves.

What should I send to a mastering engineer?

Send the clean pre-master WAV at the requested sample rate and bit depth, plus a loud reference if needed. Include notes about intentional distortion, the target vibe, references, and any alternate versions. Do not send only a crushed limiter bounce.