Producing
Parallel, Multiband, and Serial Compression: Advanced Techniques Every Producer Should Know
Why One Compressor Isn't Enough
Most producers learn compression the same way. Slap a compressor on a track, dial in a ratio, set the threshold until the meter dances, move on. That works for a while. Then you start listening to commercial records and wondering why your snare punches through the mix while yours sits there politely. Why the vocal feels glued to the track but never loud. Why the bass sounds tight at every volume.
The answer is rarely one compressor doing one job. Pro mixes lean on parallel, multiband, and serial compression to shape dynamics without crushing the life out of a performance. These techniques aren't fancy or obscure. They're the difference between a mix that sounds amateur and one that sounds like a record.
Parallel Compression: The Punch Without the Squash
Parallel compression, sometimes called New York compression, is the technique of blending a heavily compressed copy of a signal with the dry original. You get the body and weight of aggressive compression while keeping the transients and natural dynamics of the unprocessed track.
Here's the basic setup. Send your drum bus to an aux channel. On that aux, load a compressor with a fast attack, fast release, and a punishing ratio like 8:1 or 10:1. Push the threshold down until you're seeing 10 to 15 dB of gain reduction. Then blend the aux back in underneath the dry drums. Start at around 20% wet and adjust by ear.
Use a separate aux bus rather than the compressor's built in mix knob when possible. You can EQ the parallel signal separately, adding low end thump or top end air without affecting the dry signal.
The classic application is drums, but parallel compression works on almost anything with dynamic content. Bass benefits from a parallel chain when you want consistent low end without losing finger noise or pick attack. Vocals come alive with light parallel compression underneath a more transparent main chain. Even acoustic guitars and pianos can take a parallel layer for body.
The mistake most beginners make is going too loud with the wet signal. If you can clearly hear the compressed copy, you've gone too far. Parallel compression should feel like the mix got bigger without you doing anything obvious.
Multiband Compression: Surgical Dynamic Control
Multiband compression splits your signal into frequency bands and applies different compression settings to each one. Instead of compressing the whole signal when the kick hits, you can compress just the low end. Instead of taming an aggressive vocal sibilance with a de-esser, you can dynamically compress the 5 to 8 kHz range only when it gets harsh.
This is where plugins like FabFilter Pro-MB earn their keep. You can place bands anywhere in the spectrum, set independent thresholds, and even use dynamic phase to avoid the smearing that traditional multiband can introduce.
| Use Case | Band Range | Approach |
|---|---|---|
| Boomy low end | 60 to 150 Hz | 3:1 ratio, slow attack, fast release |
| Muddy low mids | 200 to 500 Hz | 2:1 ratio, medium attack |
| Harsh vocals | 2 to 5 kHz | 4:1 ratio, fast attack |
| Sibilance | 5 to 10 kHz | 6:1 ratio, very fast attack |
Where multiband really shines is on a mix bus or master. A full mix has constant frequency competition. Maybe the chorus has a synth that pushes 2 kHz too hard while the verse has perfect balance. A static EQ cut would dull the verse. A multiband compressor only acts when the chorus pushes that frequency, leaving the verse untouched.
That said, multiband compression is easy to overuse. If you find yourself reaching for it on every track, your underlying mix decisions are probably the real problem. Fix the EQ, the arrangement, or the source recording first. Multiband should be the tool that handles the last 10% of dynamic issues, not the first move.
Serial Compression: Two Compressors Are Better Than One
Serial compression means stacking two or more compressors in a row, each doing a small amount of work. Instead of one compressor pulling 8 dB of gain reduction, you might use one compressor pulling 3 dB followed by another pulling 3 dB. The total reduction is similar, but the result sounds dramatically more natural.
The reason this works comes down to how compressors behave at extreme settings. Push any compressor hard and you start hearing the artifacts: pumping, breathing, transient distortion. Spread that work across two units and each compressor stays in its sweet spot.
The best vocal chains in commercial records almost always use two or three compressors stacked, each pulling just a few dB.
The classic vocal application is a fast compressor catching transients followed by a slower compressor leveling the overall performance. Use something like the Waves CLA-76 first with a fast attack to grab peaks. Follow it with a more transparent compressor like FabFilter Pro-C 3 in opto or vocal mode, doing the slow leveling work.
The order matters. Fast compressors first, slow compressors second, when you want transient control plus leveling. Reverse the order when you want the slow compressor to set the overall envelope and the fast compressor to catch anything that still pokes through.
How These Techniques Stack in a Real Mix
Here's how this looks on a typical session. Drum bus: serial compression for glue, plus a parallel send for punch. Vocal: serial chain of fast then slow, with a touch of multiband to control sibilance. Bass: parallel compression for sustained body, plus a single channel compressor for tone shaping. Mix bus: gentle serial compression with a multiband pass to control low end on heavy choruses.
Every compressor you stack adds processing latency and can introduce phase issues. Use plugins with low latency or linear phase modes when stacking heavily, and bypass everything periodically to make sure your "improvements" are actually improvements.
None of this matters if your source recordings are weak or your arrangement is fighting itself. But once those fundamentals are solid, layered compression is what gives mixes that finished, professional weight. It's also what makes them translate across streaming platforms, where loudness normalization punishes overly compressed masters but rewards mixes with controlled, consistent dynamics. Curious how your mix loudness will translate? Check our Spotify calculator to estimate what those polished masters could earn you.
Where to Go From Here
The fastest way to internalize these techniques is to A/B them constantly. Mix a chorus with one compressor doing all the work. Then redo it with serial and parallel chains. Listen on headphones, laptop speakers, and your car. The difference will be obvious once you train your ears to hear it.
For producers building out a serious dynamics toolkit, iZotope Neutron 5 bundles multiband compression with a smart assistant that can suggest band placements based on the source material. It's not a substitute for understanding the technique, but it accelerates the learning curve.
Ready to push these techniques on your next release? You can get 7% off DistroKid here, and browse more deals on Plugin Boutique.
