Why effect order changes the sound
An effect only ever hears one thing: the signal arriving at its input. On the first slot of a channel that's the recording itself. On the second slot it's the recording as the first effect left it, and so on down the chain. Move a plug-in one slot and you change what it hears, and often what the plug-in after it hears too.
How much that changes depends on the kind of effect. Some effects do the same thing to a sound however loud it is. A static EQ that adds 3 dB at 100 Hz adds 3 dB at 100 Hz to a whisper and to a shout, and a gain plug-in turns everything up by the same amount. Engineers call these linear processes. When two of them also do the same thing from one moment to the next, with nothing like an LFO moving them, they give exactly the same result in either order. A plain reverb or delay works that way too, which is why the lab's coach says the EQ and the reverb can swap without any change.
The rest react to level, and with those the order makes a real difference. A compressor turns things down only when they cross its threshold, so anything that changes the level before it changes when it works. Distortion bends loud parts more than quiet ones, which is why a boost in front of it gives you more grit, and a noise gate opens on level, so a reverb tail in front of one holds it open for longer. Put two of these next to each other, or one of them next to a reverb that fills the gaps between the hits, and swapping them gives a different sound.
Reading the lab
The strip along the top of the lab is the chain, in signal order from left to right. Click a stage to switch it on or off, and use the two arrow buttons on each one to swap it with the stage before or after. A stage that's off passes the sound through untouched, so you can leave it where it is.
The big screen shows the whole loop. The pale shape is the sound going in and the pink shape is what comes out of the chain, drawn at the same loudness so you can compare their shapes. Whenever the compressor is on, the amber lane underneath shows how far it's turning the sound down.
The screen under it, Stage by stage, is the one to watch while you reorder things. It has a column for the sound going in and one after every stage that's switched on. The pink mark in each column is the average level of the loudest 30% of the loop, which on drums means the hits. The teal mark is the quietest 30%, the gaps between hits where a reverb tail lives, and the number above each column is the gap between the two. Each stage reacts to the column on its left, so you can see exactly what every effect was handed.
The panel next to it describes the stage you last touched, with numbers measured from the loop. When another stage is on, the lab also runs the loop with those two swapped and quotes both results, so every comparison in it comes from the sound itself. Press A/B to hear the original at the same loudness as the processed version.
A common order for an insert chain
Most single tracks start from the same order and then break it on purpose. It runs from the effects that fix problems to the ones that add space.
- Clean-up: gate, high-pass and corrective EQTake out what you don't want before anything else reacts to it. A high-pass filter removes rumble that would otherwise trigger the compressor, and a gate closes on spill and noise before compression brings them up.
- CompressionEven out the level once the problems are gone. The compressor now reacts to the sound you want, so its gain reduction follows the performance; with the rumble still in, it would also jump at every knock on the microphone stand.
- Saturation or distortionA distortion after a compressor gets a steadier level, so loud and quiet notes break up by a more similar amount. Put it before the compressor and the compressor tidies up the level the distortion leaves behind instead.
- EQ for toneBoosts and broad tone moves go here, after the compressor and the drive. A boost at this point reaches the output as you set it, and a high cut can take off the fizz the distortion just added.
- ModulationChorus, flanger and phaser go after the tone is set and before the time effects, so the delay and reverb repeat a sound that's already moving.
- DelayUsually just before the reverb. With plain plug-ins these two sound the same either way round, as the section on delay explains, so this slot only changes the sound once the delay ducks, saturates or modulates its echoes.
- ReverbReverb goes last on an insert chain, so nothing after it can squash or distort the tail. Most of the time it sits on a send instead, which keeps it out of the insert chain altogether.
A guitar pedalboard usually follows the same logic: compressor and drive pedals first, modulation in the middle, and delay and reverb at the end of the board.
| Effect | Usual place | Why there | Move it when |
|---|---|---|---|
| Gate | First | Closes on noise before anything brings it up | After a reverb, to chop the tail off on purpose |
| High-pass, corrective EQ | Before compression | The compressor stops reacting to rumble and boom | Rarely |
| Compressor | After clean-up | Reacts to the sound you want to keep | After a reverb, to bring the tail up |
| Saturation, distortion | After compression | Gets a steadier level, so it distorts evenly | Before the compressor, so it tidies up the drive |
| Tone EQ | After the drive | Shapes the distorted sound and tames fizz | Before the drive, to change what breaks up |
| Chorus, flanger, phaser | Before time effects | Delay and reverb repeat the movement | After the reverb, to move the whole wash |
| Delay | Before reverb, or its own send | Only changes the sound if the echoes are ducked, saturated or modulated | Feed its return into the reverb for echoes in the room |
| Reverb | Last, or on a send | Nothing after it squashes the tail | Into a compressor or gate, as an effect |
| Limiter | Last on the master | Sets the final ceiling | Never, apart from dither after it |
Distortion before or after reverb
Start the lab on step one. The drums go through 30 dB of drive and then the reverb, so the reverb only ever hears distorted hits. Its tail is a reverb of a gritty sound, but the tail itself decays smoothly, and between the hits the drive adds nothing: its distortion there sits more than 60 dB under the sound.
Step two swaps them. The drive now gets the dry hits and the reverb tail mixed together, so it distorts the tail too, and between the hits the distortion rises to about 35 dB under the sound. A drive also lifts quiet parts relative to loud ones, because it flattens the peaks, so the tail creeps up towards the hits: the gap in the last column of Stage by stage drops from 11.0 dB to 9.9 dB on this loop.
Both orders are useful. Drive into reverb is the safe default for guitars, synths and drums, because the space stays clean however hard you push the drive. Reverb into drive turns the tail into part of the distorted sound, and on a pad or an ambient guitar that wash can be the whole point. In the lab, turn the drive under about 18 dB and the difference almost disappears, since the drive is hardly bending the wave at all.
Compressor before or after reverb
A compressor turns the loud parts down hard and the quiet parts by little or nothing. Put it after a reverb and the reverb tail is one of the quiet parts. On the drum loop in step three, the compressor takes up to 11.3 dB off the hits and only a few decibels off the quieter tail, so the tail ends up about 7 dB under the hits instead of 11.5 dB. You hear the room swell up between every hit.
Step four swaps them back, so the compressor works on the dry drums alone. The reverb then decays at its own pace and the gap between hits and tail opens up to about 11.5 dB. That's the order you want nearly every time, and it's what you get with the compressor on the track and the reverb on a send.
The other order is a deliberate effect. A heavily compressed room or reverb under a drum kit makes the space breathe in and out with the beat, which is a form of pumping. A gate after a reverb does something else again: it lets the tail ring for a moment, then cuts it off dead, which is the gated reverb sound on a lot of 1980s snare drums.
EQ before or after distortion
A drive bends whatever is loudest. Boost the low end in front of it and the bass notes are now the loudest thing, so they break up harder. On the bass loop in step five, a 9 dB low shelf at 150 Hz before 18 dB of drive leaves the distortion about 30 dB under the sound.
Step six moves the EQ after the drive. The bass gets exactly the same extra weight, but the drive never sees it, so the distortion drops to about 39 dB under the sound. That's 8.6 dB less distortion from the same EQ setting.
So an EQ before a distortion decides what breaks up, and an EQ after it decides what the result sounds like. Guitarists use the first one all the time: an Ibanez Tube Screamer in front of a high-gain amp is popular partly because it cuts some low end and boosts the mids before the amp distorts, which keeps palm-muted chords tight. The second one is for tone. A high shelf cut after a drive takes off the fizz the distortion added, and a cut before the drive can't, because the drive makes new harmonics after it.
EQ before or after compression
The same idea applies to a compressor, because it reacts to level too. Step seven puts the 9 dB low boost in front of the compressor on the bass. The compressor hears all that extra low end and takes up to 15.5 dB off, against 10.5 dB with the EQ after it, so it turns the whole bass down harder and claws back a little of the boost: below 150 Hz holds 77% of the output in this order and 79% with the EQ after the compressor.
The usual answer is to make cuts before the compressor and boosts after it. A cut before it stops the compressor reacting to boom or rumble you were going to remove anyway, and a boost after it lands at full strength. There's a whole lesson on this one pairing, with its own lab: EQ before or after compression.
Where delay and modulation go
A plain delay and a plain reverb are both linear and don't change over time, so in series they give the same result in either order. A reverb of an echo is the same sound as an echo of a reverb. The order starts to matter when the delay isn't plain: a ducked delay turns its echoes down while the dry sound plays, a tape delay saturates and wobbles each repeat, and a reverb with modulation changes over time. Each of those reacts to what it's fed, so put the delay first and let the reverb hear the finished echoes.
Routing makes the bigger difference. With the delay and the reverb on two separate sends, the echoes stay dry and only the original sound gets the room. Send the delay return into the reverb as well and every echo sits in the same space as the dry sound, which usually blends better behind a vocal.
Chorus, flanger and phaser change over time, so their order does make a difference. Put them before the delay and reverb and the time effects repeat a sound that's already moving, while a chorus after the reverb moves the whole wash, which is wider and less focused. Watch the order of a compressor and a delay too: a compressor after a delay hears the echoes, turns the dry note down and brings the repeats up, much like the reverb in step three.
Inserts, sends and parallel chains
Everything above is about insert effects, the plug-ins stacked on a channel one after another. A send works differently: it splits off a copy of the channel and feeds it to a separate return channel, so the reverb on that return runs alongside the dry sound instead of in series with it. That's why reverb and delay so often live on sends. The compressor on the vocal channel never hears the reverb, and the reverb hears the vocal after its whole insert chain.
In most DAWs, both pre-fader and post-fader sends tap the signal after the channel's inserts; the difference is whether the channel fader comes before the send. So the EQ and compressor on a track shape what its reverb hears, which is usually what you want. If the reverb sounds too bright or too boomy, put an EQ on the return channel and leave the dry track alone. The signal flow lesson covers the whole route a sound takes through a mixer, and insert vs send compares the two side by side.
Order on a bus and on the master
A bus chain follows the same rules as a track. On a drum bus, an EQ cut, then a gentle compressor, then saturation is a common start, with any clipper near the end. On the master, the limiter goes last, because anything after it can push the peaks back over its ceiling. A clipper, if you use one, goes just before the limiter so it shaves the fastest peaks and leaves the limiter less to do.
The only thing that goes after the limiter is dither, when you export to a lower bit depth such as 16-bit for a CD. The mastering chain lesson walks through a full master in order.
How to choose an order for a chain
- Start from the common orderClean-up, compression, saturation, tone EQ, modulation, delay, reverb. It works on most sounds, and it gives you something to compare against.
- Ask what each effect should react toIf the compressor should only hear the dry vocal, it goes before the reverb. If the drive should only distort the mids, the EQ that shapes them goes before it.
- Swap the pair you're unsure aboutMost DAWs let you drag a plug-in to a new slot while the track plays. Listen to the two orders back to back, the way the lab's arrow buttons do.
- Match the loudness before you decideA new order often changes the level, and louder nearly always sounds better at first. Set the output gain so both versions play at the same loudness before you choose.
Common effect order mistakes
Compressing the reverb by accident
A reverb or delay on an insert, followed by a compressor or a limiter, gives you a tail that swells up between the notes. If you didn't mean to, move the reverb to a send or drag it after the compressor.
Boosting into a distortion when you wanted clean weight
A low or high boost in front of a saturation plug-in makes more distortion as well as more of that range. If you only wanted the tone, put the boost after the saturation and listen to the grit drop.
Gating after the reverb
A gate placed after a reverb chops the tail off whenever the level drops, which sounds like a mistake unless it's the effect you're after. To clean up spill on a drum mic, gate first and add the reverb after it.
Putting anything after the limiter
A saturation plug-in or an EQ boost after the master limiter can push the peaks back over its ceiling and undo the job it was doing. Leave the limiter last, with dither the only exception.
Choosing the louder order
Swapping two effects usually changes the output level, and the louder version wins a quick comparison. The lab matches loudness for you; in your DAW, adjust the output gain until the two orders play at the same level before you pick one.
Training your ears to hear effect order
The quiz under the lab switches on two stages and plays them in one order or the other at matched loudness, with the screens covered. Each round picks a pair: drive and reverb, compressor and reverb, or a low boost and the drive. Listen to the tail between the hits for the first two pairs and to how hard the low end breaks up for the third. Eight rounds take a few minutes, and your best score is saved in your browser.
Questions people ask
What order should effects go in?
A good default for an insert chain starts with clean-up, meaning a high-pass filter and any corrective EQ cuts, followed by compression. Saturation or distortion comes next, then EQ for tone, and the time effects go at the end with reverb last of all. On the master, the limiter goes last. Move things from there when you want a different result.
Should EQ go before or after compression?
Cuts usually go before the compressor, so it isn't reacting to rumble or boom, and boosts after it, so it doesn't fight them. In the lab, a 9 dB low boost placed in front of the compressor made it take 15.5 dB off the bass instead of 10.5 dB.
Should delay go before or after reverb?
With plain plug-ins in series it makes no difference, because both are linear: a reverb of an echo is the same as an echo of a reverb. The order makes a difference when the delay ducks, saturates or modulates its repeats, and when the two sit on separate sends, where the echoes stay dry unless you feed the delay return into the reverb.
Should distortion go before or after reverb?
Before, if you want a clean tail. With the reverb first, the distortion breaks up the tail along with the dry sound and pulls it up towards the hits. Some guitar and ambient sounds are built on that wash, so use it when you want it.
Does the order of two EQs matter?
No. A static EQ is linear, so two of them in a row give the same result in either order, and the same goes for plain gain changes. Order starts to matter once one of the effects reacts to level, like a compressor, a gate or a distortion.
Where does the limiter go?
Last on the master. Anything after the limiter can push peaks back over its ceiling. The one exception is dither, which goes after the limiter when you export to a lower bit depth such as 16-bit.