You mastered your track to -8 LUFS because it sounded huge in your studio. You uploaded it to Spotify. Now it sounds quieter than every other song on your release radar — and worse, it sounds harsher and more compressed than the tracks next to it. That's not your imagination. That's Spotify's loudness normalization doing exactly what it was built to do, and if you don't understand LUFS for Spotify, you're fighting a system you've never measured against.
Spotify doesn't play your master at the loudness you delivered it. The platform measures your track's integrated loudness in LUFS, then adjusts playback gain to hit its own target. Master louder than the target and Spotify turns you down — and the heavy limiting you applied to get there now gets heard at a lower volume, with all the artifacts and none of the loudness benefit. Master quieter and Spotify boosts you, but you've left dynamics on the table that the platform could have preserved for you.
The mistake most producers make is treating this as a numbers problem. It's not. It's a workflow problem. Here's a practical LUFS for Spotify workflow that produces masters that survive normalization and still sound competitive.
Before you touch a limiter, you need to know what number Spotify actually reads. The platform normalizes on integrated LUFS — the average loudness across the entire track from start to finish. Not peak. Not short-term. Not momentary. Integrated.
This matters because it changes how you think about dynamics. A track with a quiet 30-second intro and a loud chorus still gets normalized based on its integrated average. You're not penalized for dynamic range as long as your integrated value lands near the target. The idea that you need to crush your whole track flat to "be loud on Spotify" is backwards — what you need is an integrated average that sits where Spotify wants it.
Spotify's default target is -14 LUFS integrated. There's also a "loud" tier at -11 LUFS for Premium users who opt in, and a "quiet" tier at -19 LUFS for ambient contexts. For the majority of listeners on default settings, -14 is the number that decides your playback gain.
The second number that matters is true-peak. Spotify applies its own limiter after normalization, and that limiter can catch inter-sample peaks you didn't measure. If your master is sitting at -0.1 dBFS with ISPs above 0, Spotify's limiter catches them and you hear distortion. The platform recommends mastering with at least -1 dBTP of headroom. Set your limiter's true-peak ceiling to -1.0 dB and stop feeding the platform artifacts.
The trap producers fall into with LUFS for Spotify is staring at a LUFS meter and pushing the limiter gain until it reads -14. That produces a -14 LUFS master that sounds dead — because you hit the number by crushing dynamics instead of by balancing energy.
Here's the better workflow: pick a commercial reference track in your genre. Same era, same vibe, same instrumentation. Drop it on a channel in your DAW. Level-match it to your master so they're at the same perceived loudness — not the same peak, the same loudness. Now compare them.
If your master sounds thinner, smaller, or less punchy than the reference at matched loudness, no LUFS number will fix it. The problem is in your mix — spectral balance, transient response, low-end weight, vocal presence. A limiter can't add what isn't there. Go back to the mix. Fix the tonal balance. Then return to mastering.
Run both your master and the reference through a spectral analyzer and overlay them. Look for gaps wider than 3 dB in any band. If your low end is 4 dB lighter than the reference at 80 Hz, your track will sound weak regardless of loudness. If your high end is 3 dB hotter at 4 kHz, your track will sound harsh after Spotify turns it down. Spectral balance is the other half of the loudness equation — LUFS alone doesn't tell you whether your master sounds good, only whether it measures loud.
A master that survives Spotify normalization isn't built with one limiter at the end of the chain. It's built in stages, each stage doing one job.
Stage 1: Mix bus glue. A gentle compressor on the mix bus — 2:1 ratio, 2–3 dB of gain reduction, slow attack (30 ms or slower), auto or slow release. This doesn't make the master louder. It makes the mix hang together. Skip this and your master will sound like a collection of tracks; include it and the mix breathes as one piece.
Stage 2: Tonal balance. If your spectral analyzer showed a 3 dB dip at 2 kHz, fix it here with a broad EQ move — not a narrow cut. Mastering EQ should be gentle: wide Q, 1–2 dB moves, never more than 3 dB on any band. If you need more than 3 dB, go back to the mix. The master bus is not where you fix tonal problems — it's where you fine-tune balance.
Stage 3: Loudness. This is where the limiter comes in. Set your true-peak ceiling to -1.0 dBTP. Set your limiter ceiling to -1.0 dB. Push the gain until your integrated LUFS reads -14 — or until the limiter is doing 3 dB of gain reduction, whichever comes first. If you hit 3 dB of reduction before you reach -14, stop. Your mix doesn't have enough energy. Go back and address arrangement, individual track levels, or bus compression. Pushing the limiter harder won't add energy — it'll just squash what's there.
Stage 4: Verify. Check your integrated LUFS, your true-peak, and your crest factor. Integrated should be -14 (give or take 1 LUFS). True-peak should be under -1 dBTP. Crest factor — the gap between your average loudness and your peaks — should be 8 dB or higher for most genres. Below 8 and you've over-limited; above 12 and you may be too dynamic for the genre. Pop, hip-hop, and electronic sit in the 8–10 dB range. Acoustic and jazz can go higher.
Not every genre wants -14. An acoustic singer-songwriter track might naturally sit at -17 LUFS with wide dynamics and no limiting. That's fine — Spotify boosts it 3 dB to hit -14, and the listener gets a dynamic track at the right loudness. Don't force it to -14 with a limiter. You'll kill the feel and gain nothing.
Conversely, modern EDM and trap reference tracks often sit at -8 to -10 LUFS because that's the genre's sonic signature — smashed transients, dense energy, minimal dynamic range. If you master those genres to -14, your track sounds weak next to commercial references, even though Spotify normalizes them all to the same playback level. The reference tracks were also turned down to -14 — so at the listener's end, what they hear is the dynamics difference, not the loudness difference. Match the genre's crest factor and dynamic character. Let the integrated value land where it lands.
The rule: match your genre's reference, not the platform's target number. A -10 LUFS EDM master turned down 4 dB by Spotify still sounds like EDM. A -14 LUFS EDM master that's been artificially held back sounds timid. LUFS for Spotify is a constraint, not a goal — the goal is a master that sounds right for its genre and survives normalization without artifacts.
Mastering to -0.1 dBFS with no true-peak metering. You'll clip on Spotify's limiter. Set a true-peak ceiling and verify it.
Normalizing to -14 in the DAW and calling it mastered. Loudness isn't mastering. A -14 LUFS track can still have a muddy low end, a harsh vocal, or no transient detail. LUFS is the target; the mix is the product.
Chasing the loud tier. Some producers master to -11 hoping to land in Spotify's "loud" tier. Most listeners aren't in that tier. You're optimizing for an edge case and crushing your dynamics for listeners on the default setting who'll hear your master turned down to -14 anyway.
Believing louder masters sound better after normalization. They don't. They sound more damaged at the same loudness. The loudness war is over on streaming. The winning move is dynamics, not density.
You can't hit a target you don't measure. Run your master through a LUFS checker before you upload to Spotify. You get integrated LUFS, true-peak, crest factor, and a check against genre-specific commercial standards — the numbers that decide whether your track lands clean or gets clipped by the platform's normalizer.
If the report shows -11 LUFS integrated with -0.1 dBFS true-peak, you know exactly what to fix: back off the limiter, set a -1 dBTP ceiling, and let the dynamics breathe. If it shows -14 LUFS with 9 dB crest factor and a balanced spectrum, you're done — upload with confidence.
Stop guessing at loudness. Build the master in stages, match your genre's reference, verify your numbers, and upload a master that sounds the way you mixed it.