Streaming platforms play everyone’s music at a normalised volume, so mastering for Spotify or Apple Music isn’t about being the loudest — it’s about hitting the right loudness and staying clean. This guide explains how to master a song for streaming, what loudness to target, and how to do it free.
Short on time? Upload your mix to our free AI Mastering tool, choose a streaming target, and download a release-ready master. Here’s the full picture.
What loudness should you master to?
Loudness is measured in LUFS. Streaming services normalise playback to around −14 LUFS, so mastering much louder than that just gets turned down — and often sounds more squashed. Aim for roughly −14 LUFS for a dynamic, streaming-safe master; go a little louder (−9 to −12) only if you want more density for club or hip-hop, knowing platforms may pull it back. Our LUFS Targets reference lists the numbers for each platform.
Keep true peaks below −1 dBTP
Cap your master’s peaks at about −1 dBTP (true peak). This leaves headroom so the lossy encoding streaming platforms use doesn’t introduce distortion, and it keeps you distributor-safe. A good limiter (or an AI master) handles this automatically.
How to master a song for streaming — step by step
- Start with a good mix. Mastering polishes a mix; it can’t fix a bad one. Balance and clean it first.
- Upload to the AI Mastering tool. Use Quick Master for an instant in-browser master, or AI Reference Match to match a commercial track you like.
- Choose a loudness target — Streaming −14 for most releases.
- Compare and download a no-watermark WAV or MP3, true-peak limited and ready to upload.
Get the mix right first
The biggest gains come before mastering. If your mix is muddy or the vocal is buried, fix it first: separate and rebalance parts with the AI Stem Splitter, clean a rough vocal with the AI Audio Enhancer, or shape tone with the Online Equalizer. Then master.
AI mastering vs. a mastering engineer
For streaming singles and demos, AI mastering handles most of what you need — fast, free, and consistent. For a flagship release, vinyl, or album sequencing, a human engineer will still get you further. For everyday online releases, an AI master to −14 LUFS with −1 dBTP peaks is more than enough.
Pre-master checklist
Run through this before you master, whether you use AI or do it by hand. Most mastering problems start here.
- Leave headroom on the mix. Bounce your mix with peaks a few dB below 0 dBFS and no limiter on the master bus. A mix that is already clipped or heavily limited gives the mastering stage nothing to work with.
- Export a lossless file. Send a WAV or AIFF, never an MP3, into mastering. Keep the sample rate your project runs at (commonly 44.1 kHz or 48 kHz) and use 24-bit or 32-bit float so you keep detail.
- Check the start and end. Trim silence, make sure fades are finished and nothing is cut off.
- Know your delivery format. Most distributors accept a 16-bit or 24-bit WAV; check yours before you export the final file.
- Dither once, at the end. If you reduce bit depth, for example from 24-bit to 16-bit, apply dither on that final export only. Our guide to dithering in audio explains why.
How loudness normalisation works
Streaming platforms measure the overall (integrated) loudness of each track and adjust playback so listeners are not jumping for the volume knob between songs. If your master is louder than the platform’s reference level, it is turned down. Some platforms may also turn quieter tracks up, but policies differ and can change, so do not rely on it.
The practical result is simple: extra loudness beyond the reference gains you nothing on a normalised playlist, but the limiting you used to get there stays. A heavily squashed master ends up at the same volume as a dynamic one, just with less punch. That is why −14 LUFS is a sensible starting point rather than a ceiling to beat. For more detail, read LUFS explained: how loud should your master be.
Genre considerations
Not every style wants the same treatment. Acoustic, jazz, classical and singer-songwriter music usually benefits from staying dynamic, close to or even below −14 LUFS, so quiet passages still breathe. Dense genres such as EDM, hip-hop, trap and metal are often mastered louder for a fuller, more aggressive feel, which is where the −9 to −12 range above comes in. The trade-off is the same either way: louder means less dynamic range, and platforms may turn it down. Let the music decide, not a number.
A/B against references at matched loudness
Comparing your master with a commercial track you like is the quickest way to hear what is missing. The catch is that louder always sounds better to our ears, so an unfair comparison will mislead you.
- Pick two or three released tracks in the same genre with a sound you want.
- Load them next to your master and turn the louder one down until both play at the same perceived loudness. A loudness meter reading LUFS makes this precise.
- Switch back and forth, listening for low-end weight, vocal level, brightness and width rather than volume.
- Adjust, then re-match levels and compare again.
The AI Reference Match mode in our tool automates much of this. For a manual approach, see how to master with reference tracks.
Common mastering mistakes
- Chasing loudness. Pushing well past your target to beat other tracks, only for normalisation to cancel it out.
- Ignoring true peak. Limiting to 0 dBFS instead of about −1 dBTP, which risks distortion after encoding.
- Fixing the mix in the master. Big EQ moves at mastering usually mean the mix needs another pass.
- Judging on one system. Check on headphones, phone speakers and a car or laptop before you sign off.
- Mastering an MP3. Always master from a lossless file and export lossless for distribution.
Frequently asked questions
What LUFS should I master to for Spotify?
Around −14 LUFS, which matches Spotify’s normalisation, with true peaks at −1 dBTP. Apple Music targets a similar level. Mastering much louder just gets turned down.
How do I master a song for streaming for free?
Upload your mix to a free AI mastering tool, choose a streaming (−14 LUFS) target, and download the master. Our AI Mastering tool does this with no watermark.
Why shouldn’t I just make it as loud as possible?
Streaming platforms normalise loudness, so an over-loud master is turned down and can sound squashed. Targeting −14 LUFS keeps it dynamic and competitive after normalisation.
What is true peak and why −1 dBTP?
True peak accounts for peaks that appear after lossy encoding. Limiting to −1 dBTP leaves headroom so streaming codecs don’t distort your master.
Do I need to master every song before uploading?
Yes — mastering sets a consistent, competitive loudness and final polish, so your track sits well next to others on a playlist.
What sample rate and bit depth should I use for distribution?
Export a WAV at your project’s sample rate, usually 44.1 kHz or 48 kHz. Most distributors accept 16-bit or 24-bit files, so check yours, and apply dither only if you reduce bit depth on the final export.
Should quiet genres still be mastered to −14 LUFS?
Not necessarily. −14 LUFS is a common reference, not a rule. Acoustic, classical and ambient music can sit quieter to keep its dynamics; normalisation will handle the rest.
