Compression is the most misunderstood tool in vocal production.
Most producers know they're "supposed" to use compression, but they don't really understand what it's doing. They turn knobs, can't hear a difference, and either give up or crush the life out of the vocal trying to make something happen.
Here's the truth: compression isn't complicated—but it requires you to know what you're listening for.
In this guide, I'm breaking down compression in the simplest, clearest way possible: what it actually does, what each parameter controls, the different types of compression, and how to avoid the most common mistakes that ruin vocals.
What You'll Learn
- What compression actually does (in simple terms)
- Every compression parameter explained clearly
- Different types of compression tools (compressor, limiter, multiband, gate)
- How to get the "slammed" EDM/pop vocal sound
- Common compression mistakes and how to avoid them
- What to listen for so you can actually hear what compression is doing
What Is Compression? (The Simple Explanation)
Compression reduces the volume of the loudest parts of your vocal, bringing them closer to the quieter parts.
Think of it like this: your vocal has loud moments (when the singer belts) and quiet moments (when they whisper or sing softly). The difference between the loudest and quietest parts is called dynamic range.
Compression reduces dynamic range by turning down the loud parts. This makes the overall vocal more consistent and easier to sit in the mix without constantly adjusting the volume.
Before compression: Verse is too quiet, chorus is too loud, you're riding the fader constantly.
After compression: Vocal sits at a more consistent level throughout the song.
That's it. Compression isn't magic—it's just automatic volume control.
Why Compression Matters for Vocals
Vocals are naturally dynamic. Singers move closer and farther from the mic, shift between soft verses and powerful choruses, and emphasize certain words more than others.
Without compression, you'd have to manually ride the fader throughout the entire song—turning the vocal up during quiet parts, down during loud parts. That's exhausting and imprecise.
Compression does this automatically.
But here's the key: compression isn't just about volume consistency. When done right, it also:
Makes vocals sound more polished and professional
Helps vocals sit in the mix without getting buried or overpowering
Adds punch and presence (when using fast attack times)
Can add warmth and character (depending on the compressor)
The Problem: "I'm Turning Knobs But Can't Hear a Difference"
This is the #1 complaint about compression.
You load up a compressor, twist the threshold, adjust the ratio, mess with attack and release... and nothing sounds different. Or worse, it sounds worse but you're not sure why.
Here's what's usually happening:
Problem 1: You're Not Pushing It Hard Enough
Subtle compression is great for final mixes, but when you're learning, you need to exaggerate so you can actually hear what's happening.
The fix: Start with extreme settings:
Ratio: 8:1 or higher
Threshold: Pull it down until you're getting 6-10dB of gain reduction
Listen to what it does, then back it off to taste
Problem 2: You Don't Know What to Listen For
Compression affects dynamics (volume consistency) and transients (the attack of each word/syllable). If you're only listening to overall volume, you'll miss what compression is doing.
What to listen for:
Loud parts getting quieter - The peaks are being controlled
Overall vocal feeling more "glued" - Everything sits at a more consistent level
Transients getting softer or punchier - Attack setting controls this
Sustain and tail getting louder - As peaks come down, the quieter parts feel more present
Problem 3: You're Compensating with Makeup Gain Too Early
Many compressors have "makeup gain" to compensate for the volume reduction. If you're adjusting makeup gain at the same time as threshold/ratio, you'll mask what the compression is actually doing.
The fix: Set your compression first, then add makeup gain to match the original volume. Now A/B the bypassed and compressed versions to hear the difference.
Compression Parameters Explained
Let's break down what each knob actually does.
Threshold
What it does: Sets the volume level where compression starts working.
Think of it like this: A gate at a club. Threshold is the height requirement. Anyone above that height (louder than threshold) gets compressed. Anyone below it passes through untouched.
How to set it:
Lower threshold = more of the vocal gets compressed
Higher threshold = only the loudest peaks get compressed
Typical starting point: Set threshold so you're getting 3-6dB of gain reduction on the loudest parts.
Ratio
What it does: Controls how much compression happens once the signal crosses the threshold.
Think of it like this: If threshold is the gate, ratio is how strictly the bouncer enforces the rules.
Ratio numbers explained:
2:1 = Gentle compression. For every 2dB the vocal goes over threshold, it only goes up 1dB. (Subtle, natural)
4:1 = Moderate compression. For every 4dB over threshold, it only goes up 1dB. (Most common for vocals)
8:1 = Heavy compression. For every 8dB over threshold, it only goes up 1dB. (Aggressive, controlled)
10:1+ = Very heavy compression, approaching limiting. (Extreme control, "slammed" sound)
Typical vocal compression ratios:
Light/natural: 2:1 - 3:1
Standard: 4:1 - 6:1
Heavy/pop: 8:1 - 10:1
Extreme/EDM: 10:1 - 20:1 (or use a limiter)
Attack
What it does: Controls how fast the compressor reacts once the signal crosses the threshold.
Think of it like this: The speed of the bouncer's reflexes.
Fast attack (1-10ms):
Catches transients immediately
Smooths out the initial attack of each word/syllable
Can make vocals sound controlled and polished
Risk: Can sound dull or lifeless if too fast
Slow attack (20-50ms):
Lets transients through, catches the sustain
Maintains punch and energy
Sounds more natural and dynamic
Risk: Might not control peaks enough
Typical vocal settings:
Fast attack (5-15ms): For heavily controlled, polished vocals (pop, EDM)
Medium attack (15-30ms): Balanced—controls peaks but keeps some punch (most genres)
Slow attack (30-50ms): Preserves energy and dynamics (rock, live-sounding vocals)
Release
What it does: Controls how fast the compressor stops compressing after the signal drops below threshold.
Think of it like this: How quickly the bouncer lets go and moves on to the next person.
Fast release (50-150ms):
Compressor recovers quickly, ready for the next peak
Can sound more transparent and natural
Risk: Can sound pumpy or jittery if too fast
Slow release (200-500ms+):
Compressor holds on longer, smoothing out the overall performance
Can sound more glued and controlled
Risk: Can sound sluggish or over-compressed if too slow
Typical vocal settings:
Fast release (100-200ms): For dynamic, natural-sounding compression
Medium release (200-400ms): Balanced and musical (most common)
Slow release (400ms+): For heavily glued, controlled vocals
Pro tip: Many compressors have an "auto" release setting that adapts to the material. This is often the best starting point.
Knee
What it does: Controls how gradually compression kicks in as the signal approaches threshold.
Hard knee: Compression starts abruptly once threshold is hit. More obvious, more aggressive.
Soft knee: Compression starts gently before threshold and increases gradually. Smoother, more transparent.
Typical vocal setting: Soft knee for natural, transparent compression. Hard knee for aggressive, obvious compression.
Makeup Gain
What it does: Adds volume back after compression reduces the peaks.
Why it's needed: Compression lowers the loudest parts, making the overall vocal quieter. Makeup gain compensates for this.
How to set it: Match the compressed vocal's volume to the original (bypassed) volume, then A/B to hear the difference compression is making without volume bias.
Different Types of Compression Tools
Compression isn't just one tool—there are several types, each serving different purposes.
Standard Compressor
What it does: Reduces the volume of signals above the threshold (downward compression).
When to use it: This is your main vocal compressor. 90% of the time, this is what you're reaching for.
Typical settings for vocals:
Ratio: 4:1 to 6:1
Attack: 10-30ms
Release: 200-400ms (or auto)
Gain reduction: 3-6dB on peaks
Limiter
What it is: An extreme compressor with a very high ratio (10:1, 20:1, or even ∞:1).
What it does: Prevents the vocal from ever exceeding a certain volume level. It's a brick wall.
When to use it:
On the master bus to prevent clipping and maximize loudness
On vocals to catch stray peaks that slip through your main compressor
For the "slammed" EDM/pop sound (more on this below)
Difference from a compressor: A compressor smooths dynamics. A limiter stops peaks from going any higher. More aggressive, less transparent.
Multiband Compressor
What it is: A compressor that splits the signal into frequency bands (low, mid, high) and compresses each independently.
When to use it on vocals:
Controlling sibilance (compress just the high frequencies where S's live)
Taming boominess (compress just the low-mids without affecting presence)
Advanced tonal shaping (compress different parts of the vocal differently)
Why it's powerful: You can compress the low-mids heavily to control muddiness while leaving the highs untouched for air and clarity.
Caution: Multiband compression is advanced. It's easy to make vocals sound unnatural if you overdo it. Use sparingly.
Downward Compression vs. Upward Compression
Downward compression (standard compression): Turns down the loud parts.
Upward compression: Turns up the quiet parts (without affecting the loud parts).
When to use upward compression:
Bringing up whispered or soft sections without squashing the loud parts
Adding sustain and body to thin vocals
Parallel upward compression for thickness
Why it's less common: Most compressors only do downward compression. Upward compression requires specific plugins or techniques.
How to fake upward compression: Use parallel compression (explained below) with heavy settings on the parallel track.
Noise Gate
What it is: The opposite of a compressor. A gate reduces or mutes the signal when it drops below a threshold.
When to use it on vocals:
Removing background noise during silent sections (room noise, air conditioner, mouth clicks)
Tightening up performances by cutting off the tail of each phrase
Creating rhythmic effects (especially in EDM or electronic genres)
How to set it:
Threshold: Set just above the noise floor but below the quietest vocal parts
Attack: Fast (so it opens quickly when the singer starts)
Release: Medium to slow (so it closes smoothly without cutting off the end of words)
Caution: Over-gating makes vocals sound unnatural and choppy. Use sparingly.
How to Get the "Slammed" EDM/Pop Vocal Sound
You know that heavily compressed, upfront, loud, polished vocal sound in modern pop and EDM? Here's how you get it:
Step 1: Standard Compression (Control Dynamics)
Ratio: 4:1 to 6:1
Attack: Medium (15-30ms)
Release: Medium (200-400ms or auto)
Gain reduction: 3-6dB on peaks
This smooths out the performance and makes it consistent.
Step 2: Heavier Compression or Limiting (Slam It)
Add a second compressor or limiter after the first:
Ratio: 8:1 to 20:1 (or use a limiter)
Attack: Fast (5-15ms)
Release: Fast to medium (100-300ms)
Gain reduction: 4-8dB on peaks
This aggressively controls peaks and creates that "in your face" sound.
Step 3: Parallel Compression (Add Thickness)
Send your vocal to a parallel bus, crush it with heavy compression (10:1+, fast attack/release), then blend it under the main vocal at -15dB to -20dB.
This adds body and sustain without over-compressing the main vocal.
Step 4: Limiting on the Master
Use a limiter on the master bus to push overall loudness and catch any remaining peaks.
The result: A loud, controlled, polished vocal that sits right in your face—typical of modern pop and EDM production.
Warning: This sound is heavily processed. It works for certain genres but sounds unnatural in others (singer-songwriter, indie, jazz). Use it intentionally, not universally.
Common Compression Mistakes (And How to Avoid Them)
Mistake #1: Turning Knobs Without Knowing What You're Listening For
You adjust threshold, ratio, attack, and release... but you're just guessing.
The fix: Use extreme settings first so you can hear what each parameter does. Then dial it back to taste.
What to listen for:
Threshold/Ratio: Are the loud parts getting quieter? Is the vocal more consistent?
Attack: Are the initial transients (the start of each word) getting softer or staying punchy?
Release: Does the compression sound smooth or pumpy/jittery?
Mistake #2: Over-Compressing and Killing Dynamics
Your vocal sounds lifeless, flat, and unnatural.
Why it happens: Too much gain reduction (10dB+), too high of a ratio, or stacking multiple compressors without thinking.
The fix: Aim for 3-6dB of gain reduction on peaks. If you need more control, use multiple stages of light compression instead of one stage of heavy compression.
Mistake #3: Using the Wrong Attack Time
Too fast: Vocal sounds dull, lifeless, no punch. Too slow: Peaks slip through, vocal sounds inconsistent.
The fix:
For polished, controlled vocals: Fast to medium attack (5-20ms)
For natural, punchy vocals: Medium to slow attack (20-50ms)
Test both and listen to the difference.
Mistake #4: Not Using Makeup Gain Properly
You compress the vocal, it gets quieter, and now it's buried in the mix.
The fix: Use makeup gain to bring the vocal back to its original volume (or louder). Then A/B the bypassed and compressed versions to hear what compression is actually doing.
Mistake #5: Compressing a Bad Recording
Compression amplifies everything—including noise, room reflections, and inconsistencies.
The fix: Record properly first. Compression can't fix a poorly recorded vocal. It can only make it more consistent—for better or worse.
Mistake #6: Not Checking in Context
Your vocal sounds perfect with compression in solo, but disappears in the full mix.
The fix: Always check your compression in context with the instrumental. What works in solo doesn't always work in the mix.
Compression Order in Your Vocal Chain
Where does compression go? Here's the typical order:
Gain/trim (adjust input level)
EQ (corrective) - Remove resonances, clean up mud
De-esser (optional first stage) - Control sibilance before compression
Compression - Dynamic control
EQ (tonal shaping) - Add presence, air, warmth
Saturation/exciter (optional) - Add harmonics and character
De-esser (optional second stage) - Control sibilance after saturation
Reverb/delay (on sends) - Space and dimension
Why compression comes after corrective EQ: You want to compress the vocal after you've cleaned it up, not before.
Why compression comes before tonal shaping EQ: Compression can change the tonal balance. Final EQ adjusts the compressed sound.
How Much Gain Reduction Is Enough?
Light compression: 2-4dB of gain reduction on peaks
Transparent, natural
Good for singer-songwriter, indie, jazz
Medium compression: 4-6dB of gain reduction on peaks
Controlled but still dynamic
Good for most genres (pop, R&B, rock)
Heavy compression: 6-10dB of gain reduction on peaks
Very controlled, polished, "in your face"
Good for modern pop, EDM, heavily produced styles
Extreme compression: 10dB+ of gain reduction
Heavily processed, slammed
Good for aggressive pop, EDM, hip-hop ad-libs
General rule: More gain reduction = more obvious compression. Use only as much as you need.
Parallel Compression: The Secret Weapon
Parallel compression (also called "New York compression") is a technique where you blend a heavily compressed version of the vocal with the original.
How it works:
Send your vocal to a parallel aux track
Compress the aux track heavily (8:1 or higher, fast attack/release, 10dB+ gain reduction)
Blend the parallel track under the original at -15dB to -20dB
Why it's powerful: You get the body and sustain of heavy compression without losing the natural dynamics of the original performance.
Result: Thick, full, dimensional vocals that feel huge but still sound natural.
Common Questions About Compression
Should I use one compressor or multiple?
Both approaches work. Many engineers use 2-3 stages of light compression instead of one stage of heavy compression. This sounds more transparent and natural.
Example chain:
First compressor: 3:1 ratio, 2-3dB gain reduction (gentle control)
Second compressor: 4:1 ratio, 3-4dB gain reduction (finalize control)
Benefit: Multiple light stages = more transparent than one heavy stage.
What's the difference between a compressor and a limiter?
A limiter is just a compressor with a very high ratio (10:1 or higher). It acts as a "brick wall" to prevent peaks from exceeding a certain level.
Use a compressor for general dynamic control. Use a limiter for peak control or the "slammed" sound.
Can I compress too much?
Yes. Over-compression makes vocals sound flat, lifeless, and unnatural. If your vocal has no dynamics left, you've gone too far.
How to check: A/B your compressed vocal with the original. The compressed version should sound more polished and consistent—not lifeless.
Should I compress before or after EQ?
Both can work, but the typical workflow is:
EQ before compression (corrective): Remove resonances and problem frequencies first
Compress: Control dynamics
EQ after compression (tonal shaping): Final tonal adjustments
Do all vocals need compression?
Most do. But some naturally consistent performances might not need it—or only need very light compression.
Don't compress just because you think you're "supposed to." Compress when the vocal's dynamics need control.
Final Thoughts
Compression isn't complicated—it's just automatic volume control.
It reduces the loud parts, bringing them closer to the quiet parts, making the vocal more consistent and easier to sit in the mix.
The key is understanding what each parameter does and knowing what to listen for:
Threshold: Where compression starts
Ratio: How much compression happens
Attack: How fast it reacts
Release: How fast it recovers
Start with extreme settings so you can hear what compression is doing. Then dial it back to taste.
Most vocals need 3-6dB of gain reduction with a 4:1 ratio, medium attack, and medium release. That's your starting point. Adjust based on what you hear.