DJ Toolkit
Mastering Signal Flow

Mastering Chain Order Picker

Order matters. Put a multiband compressor before a Pultec EQ and the EQ will see a re-summation of dynamics, not the source. Put the limiter first and you've spent the "loudness budget" before you cleaned anything up. This tool lays out a defensible mastering signal chain for your genre and target medium, then explains why each stage sits where it does.

Inputs

These are starting presets, not laws. Trust the meters and your ears above the chart.

Signal flow

Left-to-right = audio path

Stage-by-stage reasoning

    How to use

    1. Pick your genre (this controls EQ suggestion + how much saturation the chain uses).
    2. Pick your source dynamics — if the mix is already at the limiter, you don't need a bus comp after it.
    3. Pick your target medium — broadcast forces a very different chain than a club master.
    4. Read the flowchart left-to-right; each block states the stage's job in one line.
    5. Build the same chain in your DAW as insert slots on a bus. Start with all plugins bypassed, then A/B them in one at a time.
    6. Compare against a reference track at matched loudness before locking it in.

    How to read the result

    • EQ first tones problems out of the way so every later stage gets a cleaner signal.
    • Compression second shapes transient behaviour; do it after spectral fixes, otherwise you compress on a frequency you then remove.
    • Saturation in the middle softens the harsh edges that compression exposes, rather than the harsh edges of the raw mix.
    • Limiter last (or clipper) so all earlier decisions reach the listener; if the limiter is anywhere earlier you "spend" loudness too soon.
    • An optional multiband belongs between the bus comp and the limiter — it is for spectral balance, not for catching peaks.

    Limitations

    Signal-chain order is style, not law. Plenty of great masters use a clipper first, or no EQ at all (mid/side only). The presets here are the most common sane starting points — they are not the only correct answer.

    Plugin choice dominates processor order. A clean linear-phase EQ behaves very differently from a Pultec-style EQ even if both sit at slot 1. The flowchart cannot capture that.

    Mastering engineers sometimes split the chain across two passes. Level-match & EQ-only pass, render, re-import, compress & limit. This tool assumes single-pass.

    Genre presets are stereotypes. "Pop" can mean anything from a Billie Eilish record (heavily compressed) to Adrianne Lenker (almost no compression). Use the genre as a guide, not a verdict.

    FAQ

    Why does EQ usually go first, not last?
    If you compress the mix first, your compressor reacts to frequencies you may then cut. EQ afterwards means your compressor sees a mix it never should have made decisions about. Doing EQ first also lets the compressor behave the same way on every play-through — it isn't chasing a moving spectral target.
    Do I really need a bus compressor on the master?
    On heavily mixed material, no — your individual busses already do "glue". On dynamic / acoustic material, a slow bus comp (10–30 ms attack, 100 ms release, ratio ~2:1, 1–2 dB GR) is the difference between 8 separate tracks and 8 tracks that sound like a record. The chart hides the bus comp for "already heavily compressed" sources for that reason.
    Why put the limiter at the very end?
    Every stage before the limiter is doing a job (subtractive EQ, transient control, glue) that gets partly erased if the limiter hits the signal too early. Once the limiter engages it has to make loudness decisions with the wrong information about what came before.
    Why does the chain look different for broadcast?
    Broadcast uses EBU R128, which caps at −23 LUFS and −1 dBTP. To meet that you usually need extra headroom — that means pulling faders down and being sparing with makeup. The chart omits the clipper option and adds a true-peak limiter with a generous ceiling.
    Where does mid/side processing belong?
    Usually right after corrective EQ. M/S EQ is for stereo-image problems (low-side mono below ~120 Hz, de-essing the sides separately), so it acts on a spectrum you've already cleaned. Putting it later means the multiband will average across both channels before your image fix, which is rarely what you want.
    Advertisement
    ============================================================ -->