Geoff Swan gave up a successful studio business to become an assistant engineer — a risk which paid off with Charli XCX’s number one album, Crash.
“My mantra for mixing is that it consists of two hours of good work and then eight hours of self‑doubt. But you need that self‑doubt! It’s integral to making sure your mix is the best it can be. My initial mix process is very fast; I get to where I want to be relatively quickly. But then I need to question it.
“I work for 15 minutes, or until I’m really enjoying the mix and think it is sounding good, and at that point I will step out of my studio and take a 10‑15 minute break. I’ll make some coffee, or engage with the kids if they’re around, or do my email. But I’ll do something completely different.
“I need to walk away every 15 minutes, or until I think the mix sounds good, because it’s like a reset, an ear break, and if you can reset, you can reduce self‑doubt. A reset allows me to check that I’m not going mad. It’s about staying focused, and not going down rabbit holes and making bad decisions. Or all the time second‑guessing myself. This is a part of my process that I have done for a long time.”
Geoff Swan’s mixing approach speaks of a singular commitment to excellence. It is undoubtedly at the heart of his meteoric rise to the top: in less than half a decade he’s gone from unknown to mixing Anne‑Marie, Calum Scott, Lady Gaga, Rudimental, Jónsi, Ed Sheeran and many more. Recently, he was also the main mixer on Charli XCX’s UK number one album, Crash. Swan’s success is rooted in a number of things, not least his extensive collection of outboard, but most of all it appears to be down to his exceptional attention to detail, and the often unusual working methods he develops as a result.
“I listen in different places in the room,” Swan explains. “The worst thing you can do is mix right in front of your speakers all the time. You can focus on your mix in a forensic manner, but it’s just not how music is consumed. I make some of my best decisions sitting on the sofa at the back, while my assistant is driving Pro Tools for a minute.
“It’s about changing perspectives. If you’re listening to NS10s all day, you’re not going to make the best decisions. So I also have Barefoot Footprint 01, Sonos, hi‑fi and Bluetooth speakers. I listen in the kitchen, or when stepping out of the room from a different space, or in the car. Listening in the car can be one of the most revealing things. If you know your car stereo it can be like cracking a code.”
Swan is talking from his mix room in the New Forest, close to the English South coast and to Southampton. He calls it The Nest, and it’s been in existence since late last Summer. This is the second version of his mix room, which he first built in 2016 when he first set up as an independent mixer following nearly five years as Mark ‘Spike’ Stent’s assistant.
“What I learned from Spike most of all was how to listen, how records should sound, trusting your instincts, and how to listen to what somebody wants, thereby establishing what you’re trying to achieve in a mix. When you’re working with somebody, you learn how they do things, but we all hear differently and we all have different tastes. Certainly when it comes to mixing and playing with balances and EQs, you have to follow your gut and your ears.
Geoff Swann: What I learned from Spike most of all was how to listen, how records should sound, trusting your instincts, and how to listen to what somebody wants, thereby establishing what you’re trying to achieve in a mix.
“In any case, you continually try to improve as a mixer. If you get stuck in doing things one way, you’re never going to get better at your craft. Trying things and experimenting with different techniques is essential. Every time I work with a new artist or producer, I learn how other people listen. If you are lucky enough to work with people multiple times, you can really build a relationship and understand how people hear things. That will inform the process for each person’s record.”
Another aspect of Swan’s career that helps to explain how he managed to become one of the world’s top mixers in just a few years is his background as a producer and studio owner. From this perspective, Swan took a big step backwards when he gave up everything he had achieved to become Stent’s assistant.
“There were times when I was doing things like cleaning SSL channel strips in the freezing cold in January, when I wondered whether I had done the right thing,” recalls Swan. “But ultimately, the learning curve was immense. To understand how records are made at that level was a learning curve towards being a better mixer. The latter was a part of my process at the time that I often fell short on, not least because I was trying to mix my own productions.”
Swan’s long journey to mixing success started, he recalls, “in my local town, Salisbury. I was playing guitar in bands during my teenage years, and often rehearsed at a government‑funded rehearsal space that also had a studio, called Base Connection. My parents were keen for me to study, so I did a Music Technology Bachelor of Science degree at Staffordshire University. By my second year I needed hands‑on experience in a studio, and ended up working at the place where I had rehearsed. This was about 2005.
“The guy who owned the studio, Pete Abbott, eventually allowed me to run the studio for him, and I ended up equipping the studio with my own gear. This is how I became a full‑time engineer. Many of the studios at that time were still working on tape, or maybe ADAT, but I knew Pro Tools, I had a Digi 002 rack, and was able to record 16 channels simultaneously, which meant that I could do things way quicker than anybody else. I was learning on the job, and it gave me the opportunity to create a business that was working.
“After running that basement studio for about three years, I met a producer called Neil Kennedy, who was running a studio in Southampton. We decided to join forces and we founded a studio called The Ranch Production House, in a farm just outside Southampton. We ran that together for five to six years. We put in a second studio and eventually built an 80m‑square live room, and we were buying gear left, right and centre, and ended up with a pretty epic collection. When I went to work for Spike, in 2012, I stepped out of that business.”
It still seems a rather bold move to give up a successful career and a partnership in a successful studio to become an assistant again. Swan acknowledges this, but explains: “When you are working with an artist at a local level, you can spend weeks making it the best it can possibly be, and it can be quite disheartening to put this thing out and to find out that only 600 people have listened to it. My quest was to work on things that people hear. It is lovely to work on something that connects with people on a broader scale.
“Also, my musical tastes changed quite dramatically. As a guitarist I was really focused on rock music: whatever it was at the time, metal, emo, punk, that is what my interest was. But as I got older, I found that I was much more interested in electronic music, in things that sonically I had not heard before, that push the boundaries and that made me question how it was done. There were no opportunities to work on that type of music at The Ranch.
“I also was very interested in what was happening in the charts, and how these tracks are mixed. All the great mixers help make things sound more visceral. When you listen to a great song that has been mixed to that standard, it connects better. It helps to increase the emotional intention of whatever the artist and producer have set out to do. And also, when you are mixing, how far can you push things? What are the boundaries? That was and remains a constant learning curve.”
Swan set up The Nest MkI in 2016. Both his old and current versions of The Nest are close to his house, and full of outboard, to the point where he works with his Pro Tools screen to the side, and has racks of outboard in front of him. The gear immediately in front of him consists of dynamics processors like the Empirical Labs Fatso and two Distressors, Locomotive Audio 14B, IGS One LA, Bettermaker Darth Limiter, a Buzz Audio REQ2.2, and saturation units like the Hendyamps The Oven, Black Box HG‑2, Overstayer Saturator NT‑02A, Thermionic Culture Vulture Super 15, SSL Fusion and the Burl Bomber B2 ADC.
Swan also has a Heritage Audio RAM System 5000 monitor controller within immediate reach, and in addition to the aforementioned Barefoot and NS10M monitors and consumer speakers, uses ProAc Studio 100 monitors powered by a Bryston 3B ST amp, and Sennheiser HD25 headphones. More dynamics options in the studio are provided by the Weight Tank WT‑Comp Rev B, POM Audio Design Boiler and PYE, two Chandler Germanium Compressors, LoopTrotter Monster, and a pair of Black Lion Audio Bluey 1176s, to name just some of the more esoteric pieces.
Swan’s reliance on outboard is against the grain in a time when most mixers have gone into the box, even former analogue diehards like Tchad Blake and Andrew Scheps, largely to be able to meet the demands of instant recalls. “My love of outboard has always been there,” explains Swan. “I started off as a guitarist, and I used to go crazy about guitar pedals. When you show me a studio with millions of things to geek out about, I’m a happy man. It was also the reason why we bought tons of outboard at The Ranch.”
Swan acknowledges that today’s plug‑ins are astonishingly good, but stresses that hardware gives him something plug‑ins can’t. “You can do everything in the box that you can do out of the box. Plug‑ins do an amazing job of emulating hardware. But I like the physical aspect of touching knobs. And I’m a big fan of saturation and running synths, guitars, drums, whatever it might be, out through bits of outboard to colour it. There’s something about hardware that you don’t get with plug‑ins. Maybe it is the randomness of it, the fact that depending on how long a piece of valve gear has been on, it’s going to sound slightly different.
“I simply get more from using hardware. Something like the Black Box HG‑2, which I use that an awful lot on my two‑bus, is a bit of a wild animal. It’s incredibly difficult to recall that piece of gear. It sounds fantastic, but getting it back to exactly where it was can be a challenge. Having said that, there’s no right or wrong way. Everything in music is opinion‑based, and so is the technical side. Who can say what is better or worse? There are people who get incredible mixes out of just a laptop with headphones and stock plug‑ins.
Geoff Swan: It’s about changing perspectives... I listen in the kitchen, or when stepping out of the room from a different space, or in the car. Listening in the car can be one of the most revealing things. If you know your car stereo it can be like cracking a code.
“It’s about finding what works for you. I just enjoy the process of working with outboard. Ultimately, I do it because I love it. I want to feel fulfilled every day in the studio, and hardware is one of the things that adds to that. When you send something out through a Culture Vulture or whatever, and you get it driving really hard, and then you bring it back in, it’s exciting. There are some amazing plug‑ins out there that do these things as well, but it’s just more gratifying for me from outboard. And I do think there is a small sonic advantage, because of the randomness.”
Still, there’s the bottom line of the instant recall requirement. Swan has a three‑pronged solution for this: a digital patchbay, printing his outboard treatments, and Session Recall. “I got the Flock Audio Patch one and a half years ago, and it revolutionised the way I mix. I was running out of I/O because of all the outboard I have, and the Flock allows me to quickly create the chains I want for each treatment. In effect, it makes outboard as versatile to use as plug‑ins.
“If I’m working on a vocal, and have it running through an outboard compressor and EQ, if I’m happy with the sound, I’ll print it. It’s a commitment. That’s something else that’s really important to me. If you find something that sounds good, commit to it, because one of the problems with having a million different plug‑ins is that you can often second‑guess yourself and you can undo a lot of good work, because you can tweak so easily. And, if I later do want to go back and change something, all our recalls are stored in Session Recall, which gives a graphical representation of every piece of hardware I own. That makes recalls very straightforward.”
Pro Tools Template
In addition to the approaches already discussed, Swan has several other mix processes that are unusual, in particular his emphasis on vocals, and the fact that he tends to mix with all tracks in. However, as with many mixers today, Swan’s process starts with a template.
Download the ZIP file of the Pro Tools Edit screenshot for a detailed view:
“My template comprises of a set of auxiliaries for my main instruments groups and stock effects that we use as a starting point. The template allows the process of setting up the session to be sped up. If people send me audio files to mix, it’s easy to drag them into the template, but if people send me a Pro Tools session, we can overlay the template. I prefer to receive Pro Tools sessions, because it allows me to choose what to change and what to retain from the rough mix.
“The first thing we do when we receive a session is that my assistant and I will look at it together, and we decide how to route things. Sometimes you get sessions with buses going into buses and you have to aware of compressors triggering other compressors and side‑chaining and so on, and it can involve quite a bit of detective work. If needed, we bring the overall level down and deal with gain staging, so there’s some headroom to work with. If something is already pushed into the red, there’s very little you can do. You’re just going to add distortion and not in a good way.
“The next stage is that my assistant takes the session to our ‘B’ ‑rig and he will clean up the vocals. It is what takes the most time in our prep process. I’m very, very keen on always having the best starting point with the vocal, so that I’m feeding all those pieces of gear with the best audio possible. Vocals are normally comped and tuned when I receive them, and we go through and spend a lot of time making everything that’s going on in the vocal musical.
“We never cut breaths, but we always automate them down, either using Clip Gain or the Trim tool. The breaths are part of the musicality of the vocals, but they need to be manageable, because they come forward as soon as the vocals hit a compressor. In general you need to be aware that as you push things up further in the mix stage, problems may become apparent, so we try to deal with them as early as possible. We’ll also manage any ASMR‑like stuff that maybe isn’t desirable, like mouth noises and so on. With pop music, you want to hear the human being behind the vocals, but at the same time it also needs to sound next‑level, almost otherworldly.
“Once the session is prepped, I start mixing. I start by looking at the vocal, but as part of the overall balance. I listen with everything in. I’ll often mute the lead vocal in and out, but I don’t have a process where I start with the kick, and then the snare, and so on. I don’t layer like that. I may solo something for five to 10 seconds to hear what’s happening, or when there’s a problem, but then I will go straight back to hearing everything in the context of the whole. Mixing is about how all elements interact with each other. It’s about looking at the whole, not the individual elements. It’s no good having the greatest kick sound in the world if it doesn’t work with everything else.
“At this point I’m mainly working with faders and a channel strip of some kind, the Brainworx bx_console SSL 4000 or the SSL Waves E‑series. Those two plug‑ins are at the heart of how I work. I do very broad strokes, nothing forensic, just working with balance and EQ to make it feel the way I want it to feel. I always have the production mix in the session, so I check all the time that I’m tonally not straying too far from where that is. Once I have a basic instrumental balance, while popping the vocals in and out, I start working on a final vocal sound, with EQ and compression. As I continue working, I keep going back and forth between the instrumental and the vocals.”
Swan illustrates his mix process by going into the details of ‘New Shapes’, the second single from Charli XCX’s fifth album, Crash. The track also features Christine and the Queens and Caroline Polacheck, and was produced by Deaton Chris Anthony and Linus Wiklund, aka Lotus IV. Swan has worked on Charli XCX’s music for a long time, mixing the whole of her Number 1 Angel and Pop 2 mixtapes, most of her third album Charli, and the whole of her fourth, How I’m Feeling Now.
Swan undoubtedly knows how Charli XCX wants things to sound, but ‘New Shapes’ provided some additional and unusual challenges because of the two additional singers. “With anything I have mixed for Charli, the production mix is the perfect roadmap for where the mix needs to go. My mix doesn’t need to deviate far from it. Mostly it was a matter of making certain elements in the production sound bolder, and dealing with three iconic vocalists, and making their vocals work together, fit in the track and each have their own space, especially towards the end, where the track gets pretty intense.
“I like to think about music as if it’s a film. You’re cutting between different scenes, even more so when you’ve got three vocalists that each have their moments. In Charli’s verse, it’s very much about her main vocal melody. Christine’s verse has many layers going on in the backing vocals, that may distract from her vocal. So I used the Wavesfactory Trackspacer plug‑in to gently key the backing vocals from the lead vocal, so when the lead vocal is featured it’s ducking the backing vocals. Caroline’s verse is the biggest scene change in the record, in terms of pace and feel. I used effects like a Valhalla Vintage Verb and a Valhalla Shimmer to form a crescendo to build back to the final chorus where everybody’s in and all the ad libs are running. It’s the most complex section of the record.
“Each of the singers have very different vocal sounds and styles. I wanted to give Charli’s vocal some pace by using compression, and also to add some spit and aggression to it. So I ran that out through my Black Lion Audio Bluey 1176 compressor, hit pretty hard, and also through my Buzz Audio EQ. I then printed that back in on the track below, on which I have the FabFilter Pro‑Q 3, for forensic stuff, and a series of sends, to a verb, a widener with the Eventide H3000, an aux track with the Weight Tank compressor, and a [Waves] H‑Delay ping‑pong delay.
“I needed a different feel for Christine, so I used the Locomotive Audio 14B, pushed and driven hard, but set more relaxed than for Charli. I also used the Buzz Audio, and then printed that back in, on a track that has the FabFilter Pro‑Q 3 and Pro‑DS, and various sends, including to reverb. Each of the singers has their own specific reverb, which is something that came from Linus.
“Caroline’s verse is much more chilled out, so the compression again is less driven. I was very much enhancing the direction that was already there, so the treatments are subtle. I sent her vocals through the Weight Tanks, a Focusrite ISA430 channel strip for EQ, and the IGS One Leveling Amplifier, which is an LA‑2A clone, and then his this gets printed on a track with the Pro‑Q 3, SSL Channel, and a handful of sends to a Vintage Verb, widener, Line 6 Echo Farm, and so on. The levels of most of these sends were heavily automated.”
Swan’s treatment of the main theme is a perfect illustration of his way of working. “The main theme is made out of three tracks that I ran through my Thermionic Culture Vulture Super 15, driven pretty hard, and also through an Overstayer Saturator NT‑02A, and then printed back in. I create different blends of saturation. I don’t mute the original track, but keep that running as well, and blend the print back in.
“There’s a Keys 1 group track that has the Waves SSL E‑channel and Plugin Boutique Stereo Savage, to add more width. I like working with width plug‑ins in general, playing with the stereo space, and being able to place things on the extremities of the stereo field. You can also use it as a dynamics tool. I quite often automate the stereo width of a certain part. In this instance, it is only coming in for the chorus.
“But for example, I will quite often add some width in the pre‑chorus, so you get the sensation of it becoming wider as it is building towards the drop on the chorus. There’s an 808 in the middle eight of ‘New Shapes’ which I ran out to the Bettermaker limiter, and I am clipping it and saturating specific frequencies, but for musical purposes, and to make sure it translated smaller speakers.
“I added a kick sample, because the drums had an AMS nonlinear verb sound, which tends to make the bottom end slightly hollow. My kick very subtly adds some additional punch to the kick. I was checking the phase alignment to make sure it was properly aligned for each hit. I also added some Waves SSL E‑channel and the Magic Death Eye stereo compressor to the original kick.
“All drums go through the All Drums bus, and also the Drum Sub A and Drum Sub B. The A bus has the ADR Compex F760X‑RS and the B the POM Audio Design Boiler. So these compressors are running in parallel to the main drum mix. They are in there for tonal reasons. Their levels are automated: they pop up and down depending on whether it is the chorus or the verse. The Boiler has distortion switches, and when they are engaged it gets a little crispy, which works really well to blend in with the drum sound.
“I ran the Compex and Boiler live, because my settings on drums tend to stay the same. These three group tracks are part of my group buses at the top, All Bass, All Keys, and so on, which are there so I can keep an eye on the metering and not slam things to hard, and they function a bit like VCAs. They’re easy if I just want to pull the vocal or something else down.
“The entire mix is running through the Mix A track at the top, which has the Waves SSL compressor, and that goes to the Buzz REQ as an insert on the master bus, lifting the top end at 16kHz little bit. It also has a saturation circuit which is adding some real drive to the bottom end. The Q‑Clone, which is not used in the mix, just gives a digital representation of what the Buzz is doing. I also have a FabFilter Pro‑L 2 on the master track, purely for listening. I did not send that to mastering.”