How to EQ in Mastering: 10 Simple Steps for Clean Mixes

Mastering EQ is about adjusting a stereo mix in tiny, deliberate increments to make it translate across every playback system without losing what the mix engineer built.

That’s the whole job. Not transformation. Not rescue. Translation.

The moves are small. We’re talking 0.25 dB to 1.5 dB. The decisions behind those moves are not small at all. Here’s how to make them correctly.

Step 1, Set Your Reference Before You Touch Anything

Don’t open an EQ first. Open a reference track first.

Load a commercially released song in the same genre into REFERENCE by Mastering The Mix or drop it onto a separate track in your DAW. Match levels to within 0.5 dBFS. Now A/B between your mix and the reference at equal loudness.

This is what mastering engineers actually do before touching a single knob. The reference tells you where the low-mids are sitting on a mix that already works, what the high end sounds like, and whether your mix is unusually thin or boxy before you’ve diagnosed anything.

Write it down. We literally keep a notepad next to the desk for this. “Reference has more air above 10kHz. Low-mids feel tighter around 300Hz.” That becomes your EQ roadmap.

Skip this step and you’re guessing. Guessing wastes sessions.

Step 2, Sweep for Problem Frequencies First

Before you shape anything, find what’s wrong.

Use a narrow Q (2.5 to 4.0), boost 6–8 dB, and sweep slowly through the spectrum. This is a diagnostic tool, not a final setting. You’re hunting for resonances, boxiness, and harshness.

The usual suspects: 200–350Hz for muddiness, 500–800Hz for honkiness, 2–4kHz for harshness that causes ear fatigue. When you find a frequency that sounds ugly and aggravated, you’ve found a candidate for a surgical cut.

We ran a dense pop mix through this sweep recently. Found a 6 dB build-up at 280Hz that was making the whole master feel like it was coming through a cardboard box. One 1.2 dB cut at a Q of 1.8 fixed it. That satisfying before/after is what keeps you coming back to this step.

Step 3, Place Corrective EQ First in the Chain

Chain order matters more than most tutorials admit.

Put your corrective EQ first. This is the surgical stage where you address the problems you found in Step 2. Use a minimum phase EQ here. FabFilter Pro-Q 4 or the Harrison AVA Mastering EQ ($89, frequently discounted to around $20) both work brilliantly for this.

After the corrective EQ, put your character or tonal EQ. That’s where you shape the overall sound of the master. Linear phase mode on a second instance works well here because linear phase introduces no phase shift, which matters when you’re doing broad, gentle boosts across a full stereo mix.

Corrective first. Tonal second. In that order, every time.

Step 4, Apply the High-Pass Filter With Restraint

Roll off below 30–40Hz with a 6 dB/octave slope. Not steeper.

A 12 dB/octave high-pass in mastering is usually overkill. It can suck energy out of kick drums and bass that lives just above the cutoff. A gentle 6 dB slope removes truly subsonic rumble, room noise, and mic handling artifacts without pulling the floor out from under your low end.

Set the cutoff at 30Hz for most music. Drop it to 25Hz for electronic or hip-hop where sub-bass is a feature. Raise it to 40Hz for acoustic music that doesn’t need the bottom.

Check the result in mono. Every time.

Step 5, Shape the Low End With a Broad Bell, Not a Shelf

80Hz is your anchor point for low-end decisions in mastering.

Use a broad bell curve: Q of 0.7 to 1.0. Adjustments here should stay within 0.5 to 1.5 dB. Any more and you’re compensating for a mixing problem, not mastering a mix.

A 1 dB boost at 80Hz with a Q of 0.7 adds weight without boom. A 1 dB cut at the same setting tightens up a mix that’s sitting heavy. Shelf filters at the low end in mastering can sound abrupt. The broad bell gives you more control over exactly where the energy sits.

Step 6, Address the Mid-Range With Surgical Precision

The 2kHz region is where presence lives. It’s also where harshness lives. They’re neighbours.

A gentle 0.5 dB cut at 2–2.5kHz with a Q of 1.2 can take the edge off a mix that sounds tiring on headphones without losing the intelligibility of vocals and snares. A 0.5 dB boost at 3–4kHz adds definition to acoustic instruments.

Keep Q values between 1.0 and 2.0 for mid-range work in mastering. Narrower than that and you’re doing mixing surgery, which is the wrong session.

Step 7, Add Air With a High Shelf Above 12kHz

“Air” means harmonic content above 10kHz that creates the sense of openness on a mix.

A high shelf starting at 12kHz, boosted 0.5 to 1.0 dB, adds that quality without introducing sibilance. iZotope Ozone 11’s Vintage EQ module handles this shelf in a way that’s hard to describe technically but sounds clean and musical rather than clinical.

We tested the same 0.75 dB boost at 12kHz with both a clean digital shelf and the Ozone Vintage EQ. The Vintage version introduced subtle even-order harmonic content above 14kHz that made the high end feel open rather than boosted. That’s the analog emulation difference, and it’s real.

Step 8, Switch to Mid/Side EQ for Stereo Problems

This is the step most tutorials mention and then abandon. We won’t.

Mid/side EQ lets you treat the centre of the stereo image separately from the sides. This is where mastering EQ gets powerful.

Common M/S moves: cut 1 dB at 200–300Hz on the sides to tighten up a washy stereo low-mid. Boost 1–1.5 dB above 8kHz on the sides to open up the stereo width without touching the mono information. High-pass the sides more aggressively than the mid, starting at 120–150Hz, to keep bass information anchored in the centre.

FabFilter Pro-Q 4 makes M/S EQ visual and intuitive. Set any band to “Mid” or “Side” mode and you’re there.

Step 9, Consider Dynamic EQ for Frequency-Specific Transient Problems

Static EQ cuts at problem frequencies affect the whole signal, even when the problem isn’t happening.

Dynamic EQ cuts only when that frequency exceeds a threshold. It’s the right tool when a mix has a vocal that occasionally gets harsh at 3kHz but sounds fine the rest of the time.

Place dynamic EQ in the corrective stage (before tonal EQ) and set gentle ratios: 2:1 to 3:1. iZotope Ozone 11’s Dynamic EQ module handles this well. Threshold settings around -20 to -25 dBFS catch transient spikes without clamping down on the mix constantly.

Step 10, Bypass and Compare at Equal Loudness

EQ adds gain. That makes the processed version louder. Louder almost always sounds better. It’s a frustrating bias that tricks engineers every single session.

Null your output level after all EQ decisions by reducing the output of your EQ chain to match the bypassed signal. Then A/B. If the EQ still sounds better at the same loudness, you’ve made a real improvement. If it suddenly sounds similar, you were just buying the volume.

We use REFERENCE or a calibrated gain-matched A/B on the master bus for this. It’s annoying to set up once. It saves you from embarrassing masters.

Pro Tips

Work in 0.25 dB increments. Dial in your move, then cut it in half. Then ask if you still hear it. If you can’t hear the difference between 0.5 dB and 0.75 dB, you don’t need 0.75 dB.

Check every move in mono. Mono collapses the stereo field and makes frequency problems obvious. A build-up at 350Hz that you might forgive in stereo becomes ridiculous in mono.

Take listening breaks. Ear fatigue at 2–4kHz is real and it makes you over-correct. After 45 minutes of EQ work, your ear will ask for cuts that don’t need to happen. Walk away for 10 minutes.

Genre shapes your target, not your taste. Electronic music lives at 60–80Hz and 8–12kHz. Jazz and acoustic music lives at 100–200Hz and 5–8kHz. Your reference track tells you what the target frequency shape looks like for the genre you’re working in.

Common Mistakes to Avoid

Boosting and cutting too much. 3 dB moves in mastering are almost always too large. If the mix needs 3 dB of correction at 300Hz, send it back to the mixing engineer. That’s a mix problem, not a mastering problem.

Using the same EQ preset for every master. Every mix has a different frequency balance. A preset that worked on a dense pop record will wreck a sparse folk record. Presets are starting points, not answers.

Ignoring phase relationships. Minimum phase EQ shifts phase as it shapes frequency. That’s fine for surgical cuts. For broad tonal boosts across the full stereo bus, linear phase mode avoids smearing the transients in your mix. Know which mode you’re using and why.

Skipping the spectrum analyser. Critical listening is the primary tool. SPAN (free from Voxengo) running in the background tells you what your ears are missing. Both together are better than either alone.