Mastering Is About Translation
A master is the version of the song that has to work everywhere. It plays on a phone speaker, a car stereo, a club system, a laptop fan-powered set of earbuds, and a pair of mastering monitors in a treated room. The job of the mastering chain is not to make the song louder or thicker. The job is to make the song translate consistently across every one of those playback systems while preserving what the mix engineer intended.
That goal shapes every decision in the chain. If a move makes the master sound bigger on your monitors but worse on a phone speaker, it is the wrong move. The chain exists in service of translation, and the order of the processors inside it matters because each one changes how the next one behaves.
The Canonical Order
There is no single correct order, but there is a canonical one that most professional engineers start from. From input to output: subtractive EQ, compression, additive EQ, stereo imaging, saturation or harmonic enhancement, limiting, and finally dither at the output if you are reducing bit depth. Every link earns its place by solving a specific problem, and the order ensures each tool sees the cleanest possible signal.
Step One: Subtractive EQ
The first move on any master is corrective. Identify and remove any imbalance the mix engineer left behind. This usually means a high-pass filter below 25 Hz to clear inaudible rumble, a gentle cut around 200 to 400 Hz to clean low-mid buildup, and occasionally a narrow notch on a specific resonance. Subtractive EQ here is conservative. If you are cutting more than 2 dB in any single band, the mix may not be ready for mastering and a return to the mix engineer is the honest move.
Step Two: Compression
Mastering compression has nothing to do with loudness. It is about controlling the macro dynamics of the song so that quiet sections and loud sections sit at the right relative balance. Ratios are gentle, typically 1.2:1 to 1.5:1, with slow attack times around 30 ms and release times tied to the tempo. Aim for 1 to 2 dB of gain reduction on the loudest moments. The compressor is not supposed to be audible. It is supposed to be felt as cohesion.
Step Three: Additive EQ
Once the dynamics are settled, the master is in a state where tonal shaping decisions reveal themselves more clearly. Now is the time for a 1 dB shelf at 12 kHz for air, a half decibel lift at 100 Hz for weight, or a wide bell around 3 kHz for presence. Additive moves at the master level should be the smallest moves you will ever make. The signal-to-noise ratio of your decision making is much lower up here, and the consequences of a wrong move are much higher.
Step Four: Stereo Imaging
Stereo work at the master stage is about widening the high frequencies and tightening the lows. A mid/side EQ that lifts the sides above 5 kHz by a fraction of a decibel can open the master, while a mono filter below 120 Hz keeps the low end punchy and translatable. Be ruthless about checking mono compatibility, because every change here has phase consequences that only a phase correlation meter will reveal.
Step Five: Saturation
A whisper of harmonic saturation across the master can do what a compressor cannot. It adds perceived loudness without reducing dynamics, and it gives digital recordings a sense of weight that pure linear processing never quite achieves. Use it sparingly. The goal is to add density that you cannot quite identify, not to colour the master in a way the mix engineer did not intend.
Step Six: Limiting
The limiter is the last creative tool in the chain. Set the ceiling at minus one dBTP to leave room for inter-sample peaks on lossy playback, and target a loudness range appropriate to the genre. Streaming platforms normalise around minus 14 LUFS integrated, so chasing minus 6 LUFS will only result in distortion without gaining loudness on the listener's end. Two to three decibels of gain reduction at the limiter is the working sweet spot for most modern productions.
Step Seven: Dither
If you are exporting at a lower bit depth than your session, dither is the final step. Apply it once and only at the very output of the chain. Multiple instances of dither, or dither applied before further processing, will undo its purpose and introduce noise instead of removing quantisation artefacts.
Listening for the Whole, Not the Parts
The temptation when learning mastering is to evaluate each processor in isolation. Resist it. Once the chain is built, your only job is to compare bypass against active on the entire chain and ask one question: is the master more translatable than the mix it started from? If the answer is yes, ship it. If the answer is "louder but not better," dial back.
Precision tools earn their place at this stage of the workflow. The narrowest moves come last, and that is where a transparent dynamic EQ like TRYKZ EQ pays for itself by keeping corrections invisible while the rest of the chain does the audible work.
The best masters do not announce themselves. They simply sound right on whatever you play them through, and the listener never has a reason to think about the engineering at all.
A Final Word on Reference Tracks
No mastering session should run without reference tracks loaded at matched loudness. Match the integrated LUFS of your reference to your master before comparing, otherwise the louder one will always sound better. The job is to learn what professional masters in your genre sound like at the same loudness as yours, and to get yours closer to that standard with every pass.