Mastering Guides - Loudness, Chains & Streaming-Ready Masters

To master a song, you take your final stereo mix and make three things happen: you balance the overall tone so it sounds right on every system, you control the dynamics so it stays punchy and consistent, and you raise the loudness to a competitive level while keeping the true peak safely below the ceiling. In practice that means a short chain - corrective EQ, gentle compression, tonal shaping, stereo control, then a limiter - pushed to around -14 LUFS integrated with a true peak near -1 dBTP for streaming, checked against reference tracks, and exported as a high-resolution WAV. That is the whole job in one breath. The rest of this guide unpacks each step so you can do it well, whether you master by hand, with AI, or by hiring a pro.

What is mastering, and what is it not?

Mastering is the final creative and technical step between a finished mix and a released track. It works on the stereo file as a whole, not on individual instruments. A mixing engineer moves the kick up, tames a harsh vocal, and pans the guitars. A mastering engineer takes that completed stereo bounce and treats it as one object: shaping its overall balance, glueing its dynamics, setting its loudness, and preparing clean deliverables for every platform.

The clearest way to understand the boundary is to read mixing vs mastering as two different jobs with two different toolsets. Mixing is where you fix problems inside the song. Mastering is where you polish the finished result and make it translate everywhere - car speakers, phone, earbuds, club system, laptop. Mastering is not a rescue mission. It cannot un-muddy a bad low end, separate two instruments fighting for the same frequency, or fix timing. If something is broken, it belongs back in the mix. Master a strong mix and you get a strong master. Master a weak mix and you get a louder weak mix.

What does the mastering chain look like, and what order do the plugins go in?

Signal order matters because each processor reacts to what comes before it. A limiter hears whatever the EQ and compressor hand it, so the sequence changes the result. A reliable default chain looks like this, top to bottom:

• Corrective EQ - surgical cuts to remove rumble, boxiness, or a harsh peak before anything else amplifies it.
• Compression - one gentle pass, usually 1 to 3 dB of gain reduction, to glue the mix and even out level.
• Tonal (musical) EQ - broad, wide moves for character: a touch of air up top, a little weight down low.
• Stereo and saturation - subtle width and harmonic color to add depth and perceived loudness.
• Limiter - the last stage, catching peaks and setting final loudness and the true peak ceiling.

You do not need every link every time. A clean modern mix might only want EQ and a limiter. For a full walkthrough of why each stage sits where it does and how to adapt the order, see the mastering chain and plugin order. The rule to remember: corrective before musical, dynamics before loudness, limiter always last.

How loud should a master be for streaming?

Loudness is measured in LUFS (Loudness Units relative to Full Scale), which tracks perceived volume over time rather than raw peaks. Every major streaming platform normalizes playback to a target, then turns tracks up or down to match. Here are the real numbers you are aiming at:

• Spotify, YouTube, Tidal, Amazon Music - around -14 LUFS integrated.
• Apple Music - around -16 LUFS integrated.
• True peak - keep it at or below -1 dBTP everywhere to avoid inter-sample clipping after lossy encoding.

These targets are the whole reason loudness normalization exists. Deliver a track at -8 LUFS and Spotify simply turns it down to match -14, so all that crushing bought you nothing but lost dynamics. Deliver at -20 LUFS and it gets turned up. For the full breakdown per service, read what LUFS to target for streaming, and if you want to understand the ceiling specifically, true peak explained covers why -1 dBTP protects you from distortion that only appears after conversion to MP3 or AAC.

Why is louder not better anymore?

The loudness war made sense when radio and CD had no normalization: the louder track jumped out. Streaming ended that. Because every platform now matches perceived loudness at playback, an over-compressed master does not sound bigger to the listener - it sounds turned down and flat, with the transients and dynamics already squashed out of it. The track that wins under normalization is the one that sounds open, punchy, and full at the target loudness, not the one that measured highest before normalization pulled it back.

If your own songs feel small next to commercial releases even after you push the level, the culprit is usually mix balance and arrangement, not the limiter. Why your song sounds quiet digs into perceived loudness versus measured loudness, and how loud a master should be gives you a target-first way to think about level.

What does each processor do at the master stage?

The tools look familiar from mixing, but at the master stage you use them with a much lighter touch because they act on everything at once.

EQ

Master EQ is about broad tonal balance, not surgery. Wide, gentle moves of 1 to 2 dB shape the overall curve: a high shelf for air, a low shelf for weight, a narrow cut only when a specific resonance pokes out. If you find yourself making big corrective moves, the fix belongs in the mix.

Compression

Master bus compression is glue. Slow attack, auto or medium release, a low ratio around 1.5:1 to 2:1, and only 1 to 3 dB of gain reduction. The goal is cohesion and a slightly more consistent level, not obvious pumping.

Stereo and dynamics

Subtle widening can add space, but keep the low end mono for a solid, phase-coherent bass. Watch your dynamic range: a master that is too flat feels lifeless, while one with healthy transients feels alive even at a lower measured level.

The limiter

The limiter is the most important and most misused tool in mastering. It catches the loudest peaks and lets you raise overall level, setting both your final loudness and your true peak ceiling. Aim for a few dB of gain reduction on the loudest moments, not constant heavy limiting. What does a limiter do explains ceiling, threshold, and release in plain terms so you can push level without smearing your transients.

How do you use reference tracks?

A reference track is a professionally mastered song in a similar style, loaded next to your master so you can compare tone, loudness, and width honestly. Your ears adapt fast and drift toward whatever they have been hearing, so an external anchor keeps you objective. Level-match the reference to your master first, because the louder of any two tracks almost always sounds better for the wrong reasons. Then compare: is your low end bigger or smaller? Are your highs duller? Is the reference wider? How to use a reference track walks through choosing good references and matching them properly so you are comparing craft, not volume.

What headroom and gain staging do you need before mastering?

The single most common mistake is handing off a mix with no room to work. If your mix bus is already slammed against 0 dBFS, the master has nowhere to go and every processor is fighting a clipped signal. Leave headroom: aim for mix peaks around -6 to -3 dBFS and remove any limiter or clipper from your mix bus before you bounce for mastering. Clean gain staging through the mix also means your master starts from a controlled, undistorted signal. Gain staging and headroom covers exactly how much room to leave and how to set levels so mastering has something to work with.

How do you export a mix for mastering?

Bounce a stereo WAV (or AIFF) at the highest quality your project allows. Use 24-bit depth so you preserve dynamic range and quantization noise stays inaudible, and keep the native sample rate of your session, whether that is 44.1, 48, or 96 kHz - do not upsample or downsample on the way out. Never export an MP3 to master from; lossy compression bakes in artifacts you cannot remove. Leave the headroom mentioned above and turn off master bus limiting. The full checklist, including dithering and file naming, is in how to export a WAV for mastering.

Do you master differently for each platform?

For most releases, one well-made master at roughly -14 LUFS integrated with a -1 dBTP ceiling covers Spotify, YouTube, Tidal, and Amazon cleanly, and Apple Music will normalize it down to its -16 target without harm. You do not need a separate file per service in most cases. Where platform-specific attention pays off is at the extremes and for specific destinations. Mastering for Apple Music covers Sound Check and the Apple Digital Masters approach, mastering for YouTube handles video loudness, mastering for SoundCloud deals with its lighter normalization, and if you work in spoken word, how to master a podcast targets the different loudness standards that voice content uses.

Does mastering change by genre?

The chain stays the same, but the targets shift. Bass-heavy, transient-driven styles need careful low-end control and a limiter that respects punch, while acoustic and vocal-forward music prioritizes dynamics and natural tone. If you produce beats, the sub and the drums decide everything, so genre-specific guides help: how to master a trap beat for booming 808s and crisp hats, how to master a drill beat for sliding sub-bass, and how to master amapiano for its log-drum-driven low end and spacious groove. Start from the same signal chain, then tune the low-end balance, saturation, and loudness to what the genre expects.

How do you master in your DAW?

Every DAW can master a track with its stock plugins - the workflow matters more than the brand. Set up your chain on the master bus, work in the order above, and meter loudness and true peak as you go. If you want step-by-step, DAW-specific instructions, read how to master a song in FL Studio and how to master a song in Ableton. Both cover routing, the stock chain, and metering so you can get a competitive result without buying anything. For the fundamentals that apply to any DAW, how to master a song is the core method.

Should you master yourself, use AI, or hire a pro?

All three are valid, and the right choice depends on your track, your budget, and your ears.

• DIY gives you total control and teaches you the most, but it takes practice and honest monitoring to avoid over-processing. It is the best way to learn and the cheapest at volume.
• AI mastering analyzes your mix and applies a balanced chain tuned to streaming targets in seconds. It is fast, consistent, and effectively free to audition, which makes it ideal for demos, drafts, releases on a budget, and comparing options before you commit. A good AI master gives you a loud, clean, platform-ready file instantly, and you can iterate as many times as you like.
• A professional engineer brings experienced ears, high-end monitoring, and creative judgment that matters most on flagship releases. It is the priciest and slowest option, and it shines when the stakes are high.

Many artists use these together: draft with AI to hear the track at full loudness immediately, learn from what it does, and reserve a pro for the single that matters. If your mix is solid and you leave proper headroom, an AI master will get you a release-ready result today.