Best way to get loudness is through saturation in small amounts.

Pretty much every sound in my tracks have a saturator with soft clipping on which is acting as a limiter. I group all of my tracks into synths, basses, sub, drums etc. As well as saturating each track individually, I'll have one on the group channel too and feed that into a saturator a few db, just to squash things down a little. (I'll usually do this step last and fine tune it when I'm doing my final mixdown. I end up turning things up in the group that I feel aren't loud enough and they're being fed into the saturator too hard).

I'll also get my kick and snare sitting at 0db using saturation. I tried this for the first time with my last few tracks and it made mastering really easy for me. I'd make sure that the saturator was trimming a few db off my kick and snare without affecting the sound.

The more 'sausaged' your mix is before you it reaches your mastering chain, the more loudness you can get from it. You have to be really careful with saturation cause it's easy for your mix to sound completely flat and squashed but when it's used in small amounts, it's really helpful for getting loudness.

Then it's just a matter of sidechaining the kick and snare to everything else in the track or cutting small gaps in the track so he can slot the kick and snare in and have absolute silence to minimise clipping

I've read and re-read this sentence a dozen times but I don't quite understand it... can you please eli5?
[/quote]

For sure! So once he's made his two mastering chains, he'll have a track of all his synths, basses, sub fx, and percussion (i'll just refer to this as the 'track') and then he'll have a second track with just his kick and snare on.

Just to make this easier to explain, let's say he's bounced these two tracks to audio and put them into a fresh project in Logic.

So he's got 2 audio tracks in his project, his kick and snare and 'track'. Everytime a kick and snare plays, he'll cut a tiny section out of the 'track' so that he has complete silence on the initial hit of the kick or snare, kind of like an extreme sidechain. If he were to cut this extremely accurately so that his kick or snare aren't playing at the same time as the 'track' or overlapping, this would result in the master not clipping, in theory. (And because these two tracks have already been mastered, the waveform would still look totally brick walled because nothing is overlapping, he's just slotting the drums into the silence that he's cut out of the 'track')

Since the snare has a fairly long tail on it, if he were to cut out a gap in the track for as long as the snare tail decays, it would sound awful and the 'track' would suddenly jump in volume when the tail has decayed completely. He's probably cut a tiny bit out for a few milliseconds where the snare transient hits leaving the tail of the snare overlapping with the 'track'. By doing this, his limiter doesn't have to work as hard, it only has to limit where the tail of the snare overlaps with the 'track' as opposed to having a huge transient to deal with.

Hope this clears it up, I'm really bad with explaining things haha