How to Normalize Audio for Spotify at -14 LUFS (and Why It Works)
Spotify plays every track and episode at about -14 LUFS. Upload something quieter and Spotify turns it up. Upload something louder and Spotify turns it down. Either way, Spotify decides — unless your file already sits at the target. This guide shows how to land at -14 LUFS yourself, and what happens if you do not.
The fast road: choose the Streaming & YouTube preset in the loudness normalizer. The guide below explains the why behind the number.
Why -14 LUFS?
Streaming services picked -14 LUFS as a shared middle ground: loud enough to feel full on earbuds, quiet enough to leave headroom so peaks do not clip when the file is re-encoded for streaming. Spotify documents this playback normalization for artists and podcasters. YouTube uses the same -14 reference, which is why one preset covers both — see the YouTube loudness guide for the video side.
The number also ended an arms race. In the CD era, mastering engineers made each release louder than the last, crushing dynamics to win a loudness contest that listeners never asked for. Playback normalization removed the prize: once the platform turns everything to the same loudness anyway, a crushed master gains nothing and loses its punch. -14 LUFS is the level the industry settled on where a dynamic mix and a loud mix play back at the same volume — and the dynamic one usually sounds better.
How Spotify Treats Your Upload
| Your file measures | What Spotify does | What the listener hears |
|---|---|---|
| -20 LUFS | Turns it up about +6 dB | Your noise floor comes up too |
| -14 LUFS | Leaves it alone | Exactly what you mastered |
| -8 LUFS | Turns it down about -6 dB | Fine, but your loud master bought you nothing |
The lesson: mastering louder than -14 buys nothing on Spotify, and mastering quieter costs you clarity. The target is the destination, not a minimum.
Two details are worth knowing. First, Spotify applies this as a plain gain change — it does not compress or re-master your audio. A quiet, noisy recording turned up 6 dB is the same recording with its noise 6 dB louder. Second, Spotify's normalization is applied per track or episode, so in a podcast feed it keeps episodes consistent with each other only if they started consistent in character; wildly different mixes still sound different, just at matched average loudness.
A Worked Example: One Song, Three Masters
Take a finished mix exported three ways and watch what a listener actually gets:
| Master | Integrated | True peak | Spotify's gain | Playback result |
|---|---|---|---|---|
| Quiet master | -19 LUFS | -3.0 dBTP | +5 dB | Peaks hit +2 dBTP territory on decode — risky, and hiss is up 5 dB |
| Target master | -14 LUFS | -1.0 dBTP | 0 dB | Exactly the master, peaks safe |
| Loud master | -9 LUFS | -0.2 dBTP | -5 dB | Same loudness as the target master, but the mix was squashed to get there |
Read the first row again: turning a quiet file up at playback can push its peaks over the top, because gain does not know where the peaks are. That is why the job belongs in your studio, with a true-peak ceiling in place, and not in the player's playback adjustment. The true peak guide explains the ceiling in full.
How to Normalize to -14 LUFS
- Finish the mix or episode first; loudness is the final step.
- Measure the integrated loudness with a BS.1770 meter (the LUFS explainer defines the terms).
- Normalize with a two-pass tool: measure, apply the exact gain, hold true peak at or below -1 dBTP.
- Re-measure the result and confirm -14 LUFS ±1 before you upload.
The normalizer on the home page does all four steps and shows the before/after report. Podcasts headed to Apple Podcasts as well should use -16 instead — that trade-off is covered in the podcast normalization guide.
For music, the same steps apply per track. If you are releasing an album, normalize track by track to -14 and then listen to the album in order: album playback keeps relative track levels, so a ballad you forced to -14 will sit louder against its neighbours than you intended. Many engineers master albums as a unit and let individual tracks land a little under -14, accepting Spotify's gentle turn-up rather than breaking the album's flow. For singles and podcast episodes, hit -14 exactly.
Spotify's Loudness Settings — and Why You Still Master to -14
Spotify's app has a volume-level setting (Quiet, Normal, Loud) that shifts playback loudness for that listener. It is tempting to master for the Loud crowd or the Quiet crowd. Do not. The setting moves every track together, so the relative picture never changes: your master is still compared against everyone else's -14. Mastering to the documented reference is the only move that works for all listeners at once.
Similarly, listeners can turn normalization off. Then your raw file plays as-is, next to raw files mastered by people who did target -14. A file mastered to the reference behaves well in both worlds; a file mastered at -8 to "win" with normalization off just sounds harsh and clips earbuds at volume.
Track Mode and Album Mode
One more Spotify behaviour matters if you release music. When a listener plays a single track or shuffles, Spotify normalizes each track on its own. When they play an album straight through, Spotify keeps the album's internal balance and adjusts the album as a whole, so the quiet intro track stays quiet against the closer — the way the artist sequenced it.
This is good news and a warning. Good news: you do not have to flatten an album to protect its dynamics; the platform respects the journey. Warning: a track ripped out of an album context — into a playlist, a shuffle, a podcast-style feed — gets judged on its own integrated number. If your album's opening track sits at -20 because the album arc demanded it, that track alone in a playlist will play quiet. Singles strategy exists for a reason: anything you expect to travel alone should stand at -14 on its own feet.
After Release: How to Check What Spotify Did
You cannot read Spotify's applied gain directly from the listener app, but you can verify the outcome. Play your release in Spotify at your usual volume, back to back with a reference track you know sits at -14. If you reach for the volume in either direction, something is off — usually your file's loudness range rather than its integrated number. Then check the mundane causes first: an old, quiet release still cached in a playlist, or a distributor copy that predates your loudness routine.
For podcasts, the same test in your regular podcast app queue tells you more than any meter: your episode against a network show, same headphones, same volume setting. Loudness work is finished when that test becomes boring — episode after episode, nothing to adjust.
One File, Two Platforms?
A podcast episode usually ships to both Apple (-16) and Spotify (-14). Two LUFS apart is close enough that many shows publish the -16 master everywhere and accept Spotify's gentle +2 turn-up. If your audience is mostly on Spotify and YouTube, master at -14. What you should not do is publish far from both and let each platform guess differently.
The same logic covers other streaming services: most music platforms cluster around -13 to -16 LUFS playback references, so a -14 master is never far from any of them. Broadcast is the outlier — European delivery at -23 LUFS under EBU R128 is a different master, not a compromise.
Common Mistakes
Mastering as loud as possible "for impact." On Spotify, impact past -14 is deleted by the turn-down. What survives is the distortion and flatness you added to get loud.
Judging loudness by the waveform. A fat waveform shows peaks, not perceived loudness. A sparse acoustic track can look small and measure -13 LUFS; a dense, quiet-in-feel mix can look huge and measure -18. Measure, do not eyeball.
Normalizing the MP3 you downloaded back. Work from your master file. Normalizing an already-encoded copy stacks encode artifacts and can push peaks in ways the report will show but your ears will regret.
Forgetting intros, ads and bonus clips. Spotify normalizes each item, so a screaming ad read inside a -14 episode still screams within the episode. Integrated loudness is a whole-file average; internal balance is your job in the mix.
Re-uploading to fix loudness alone without checking. If an old release is far off, a corrected re-upload through your distributor is legitimate — but verify the new file measures on target first, or you will pay the re-delivery cost twice.
Troubleshooting
My track sounds quieter than others on Spotify despite measuring -14. Check the loudness range. A dynamic track at -14 has verses well below -14, and listeners compare the verse they are hearing now. Dense, low-range tracks feel louder at the same integrated number — that is physics, not a bug. If it bothers you, a touch more compression in the mix is the fix, not a hotter master.
Spotify turned my quiet track up and now it hisses. The turn-up raised your noise floor with everything else. The real fix is a cleaner or louder original: re-normalize from the master at -14 with a peak ceiling so the platform leaves it alone.
My podcast measures -14 but Apple listeners say it is loud. It is — by 2 LU against the -16 convention. Either ship the -16 master everywhere, or keep two masters. Both are defensible; guessing is not.
What This Costs on LoudNorm
The normalizer on this site has no watermark and no jingle on any plan, including the free one — some rival tools stamp free output, this one never does. The free plan covers 60 minutes of audio a month, which is several podcast episodes or a handful of tracks. Starter ($9) covers 10 hours a month, and Pro ($19) covers 30 hours a month and suits catalogues and batches. Every plan runs the same two-pass measurement and produces the same before/after report; the paid tiers buy time and batch, not accuracy.
Frequently Asked Questions
Does Spotify normalize podcasts and music the same way? Yes — the same -14 LUFS reference applies to episodes and tracks, which is why one streaming preset serves both.
Should I use -1 dBTP or -2 dBTP true peak for Spotify? Spotify's guidance pairs -14 LUFS with -1 dBTP. Use -2 dBTP if your distributor or another destination asks for it; the loudness difference is negligible and the extra headroom never hurts.
What file format should I upload? The highest quality your distributor accepts — WAV or FLAC where offered. Spotify encodes from what you send; a good master encoded once beats an MP3 encoded twice. Normalize the master, then let the distributor's encode be the only one.
Does loudness affect discovery or playlisting? There is no evidence Spotify rewards or punishes any loudness in ranking. What loudness changes is the listener's first ten seconds: a track that starts at the right volume keeps people listening instead of skipping, and that alone is reason enough to get it right.
I release on YouTube too — same file? Yes. YouTube's reference is also about -14 LUFS, and its loudness behaviour mirrors Spotify's for loud content. One -14 master serves both.
References
- https://artists.spotify.com/ — Spotify for Artists, documenting -14 LUFS playback normalization and the -1 dBTP peak guidance.
- https://support.google.com/youtube/answer/1722171 — YouTube Help (Google), on audio quality and loudness for uploads.
- https://www.itu.int/rec/R-REC-BS.1770 — ITU-R BS.1770, the measurement standard behind LUFS.