
Audio headroom is the gap between the loudest peak in your signal and the maximum level the system can hold — in the digital world, that ceiling is 0 dBFS (decibels relative to full scale). If your track's tallest peak hits -6 dBFS, you have 6 dB of headroom: six decibels of room before the signal runs into the ceiling and distorts. Headroom is simply that safety margin, and leaving some is one of the cheapest ways to keep a mix clean.
Every digital audio system has a hard ceiling. Samples are stored as numbers with a fixed maximum, and 0 dBFS is the point where those numbers run out of room. Push a peak past it and the system can't represent the value, so it flattens the tops and bottoms of the waveform instead. That flattening is audio clipping, and it adds harsh, brittle distortion that no amount of later processing fully removes.
Headroom is the buffer that keeps you away from that edge. It isn't a setting you switch on — it's just the distance you've chosen to leave between your peaks and the ceiling. The question is never whether you have headroom (your peak level always sits somewhere below or at 0 dBFS), but whether you've left enough of it for what you still need to do.
It's tempting to think a peak sitting right at 0 dBFS is "using the full signal" and therefore better. In practice, running that hot causes three separate problems.
You give clipping no margin for error. Audio is rarely finished the moment you set a level. You add an effect, automate a fade, bounce to a new format — and any of those can nudge a peak upward. If your loudest moment already touches 0 dBFS, the smallest change tips it over into distortion. A few decibels of headroom absorbs those surprises.
You leave no room for mastering and processing. Mixing and mastering are additive. EQ that boosts a frequency raises the level there. Compression changes the peak-to-average relationship. Summing many tracks together stacks their energy. If your mix bus is already slammed against the ceiling when you hand it off, every one of those stages has to fight the limiter instead of shaping the sound. Headroom on the mix bus is what gives the next stage somewhere to work.
You expose yourself to inter-sample peaks. This is the subtle one. A digital file stores discrete sample points, but playback reconstructs a smooth continuous wave between those points — and that reconstructed wave can rise higher than any stored sample. A file that reads -0.1 dBFS on an ordinary sample-peak meter can actually overshoot past 0 when a DAC or a lossy encoder rebuilds it, producing distortion your meter never warned you about. A true-peak meter (measured in dBTP, decibels true peak) predicts those overshoots. Leaving headroom is how you stay clear of them.

There's no single correct number, because the right amount depends on where you are in the process. Two rules of thumb cover most cases.
While mixing, aim for a mix bus that peaks around -6 to -3 dBFS. That gives your mastering stage real room to add level, glue, and tone without immediately hitting the ceiling. It also keeps individual tracks from summing into a clipped master. You're not chasing a precise target here — you just want obvious space above the loudest peak, not a mix that's already flat against 0.
On your final export, set a true-peak ceiling of -1 dBTP. This is the widely accepted convention for a finished master, and it's about inter-sample peaks rather than caution for its own sake. Keeping true peaks at or below -1 dBTP leaves roughly a decibel of margin for the overshoots that appear during reconstruction and lossy encoding, so the file that sounds clean on your system stays clean after a streaming platform re-encodes it. Some platforms ask for this explicitly — Spotify's documentation requests true peaks below -1 dBTP — but even where it isn't stated, it's a safe default.
Note that these two numbers do different jobs. The -6 to -3 dBFS mixing figure is working headroom, giving later stages room to operate. The -1 dBTP export figure is a protective ceiling on the finished file. One is about process; the other is about the delivered result.
These three terms get tangled together constantly, because they all involve levels and decibels. They measure different things.
A quick way to keep them straight: headroom looks at the space above your peaks, dynamic range looks at the distance between your quiet and loud parts, and loudness is the perceived average of the whole thing. You can have generous headroom and tiny dynamic range (a quiet but heavily compressed track), or slam your peaks to the ceiling while still measuring quiet in LUFS (a sparse mix with one sharp transient). They move independently, which is exactly why it helps to treat them as separate ideas.
The practical link between them is worth knowing: the loudness war of chasing a bigger LUFS number usually means limiting peaks harder to raise the average — which eats both headroom and dynamic range at once. Because streaming platforms normalize playback to a loudness target anyway, that trade often buys nothing and costs the life of the mix. Leaving headroom and letting normalization set the level is the calmer path.
You don't need studio software to check or set headroom. On vocalcut.com, every tool runs entirely in your browser — your audio never leaves your device, and there's no account or upload step (the processing engines download to you, not your files up to a server).
When you're ready to finalize a track, the in-browser mastering chain includes limiting with a true-peak-aware output ceiling, so you can set that -1 dBTP target on export and see your peaks respected rather than guessing. If you just need to move a finished file to a consistent level without reshaping it, the volume normalizer measures the track and applies gain to hit your target while keeping peaks in check. For a step-by-step walkthrough of getting a file to a target level cleanly, see how to normalize audio volume.
Headroom is a small habit with an outsized payoff: leave a little room above your peaks, protect the export with a -1 dBTP ceiling, and most clipping problems never happen in the first place.
How much headroom should I leave when mixing? Aim for a mix bus that peaks around -6 to -3 dBFS. That leaves your mastering stage real room to add level and tone without immediately hitting the 0 dBFS ceiling, and it keeps individual tracks from summing into a clipped master. The exact number matters less than having obvious space above your loudest peak.
What does -1 dBTP mean and why is it recommended? dBTP stands for decibels true peak — a measurement that predicts inter-sample peaks, the overshoots that appear when a file is reconstructed for playback or re-encoded to a lossy format. Setting a -1 dBTP ceiling on export leaves about a decibel of margin for those overshoots, so audio that looks clean on a normal sample-peak meter stays clean after encoding. It's an accepted convention rather than a universal rule, but a safe default.
Is headroom the same as dynamic range? No. Headroom is the gap between your loudest peak and the ceiling (0 dBFS). Dynamic range is the span between your quietest and loudest parts. You can have plenty of headroom and very little dynamic range, or almost no headroom yet a wide dynamic range — they measure different things and move independently.
Will leaving headroom make my track too quiet? No, because loudness and headroom are separate. Headroom controls how close your peaks get to the ceiling; perceived loudness in LUFS is set later by normalization or mastering gain. A track with sensible headroom can still be brought to any target loudness — and since streaming platforms normalize playback anyway, leaving headroom rather than maxing peaks usually sounds better at the same final volume.