Jonathan Farber Via Unsplash
Mixing Basics: How to Mix a Song from Start to Finish
Starting a mix is never easy. Whether you’re working on your own song, mixing for a client or simply practicing your craft, the hardest part of mixing is often knowing where and how to start.
The internet is full of advice about this critical stage of song making – some telling you to “trust your ears”, which can feel too vague and unhelpful, and some using charts and diagrams that turn mixing into a formula where every step must be done in the right order. Neither of these approaches are technically wrong, but I think the ideal system should include both: enough direction for a clear path forward, and enough freedom to experiment and figure out what the song really needs.
This article is meant as a framework to start building your own system. The steps are by no means strictly chronological, and the advice is purposefully open-ended. Each song needs something different to bring it to life, and a good system reminds you of your options, instead of telling you which ones to use. We’ll look at how to prepare your project, identify key sounds, clean up and process every element, add depth and movement, and finish with a cohesive and balanced mix.
The goal isn’t to teach you what to do – it’s to show you how mixing engineers approach a mix, with a system that is both repeatable and adaptable. That being said, there is a starting point that all mixes have in common – the first listen.
1. Analytical Listening: Understanding the Vision
Before you start mixing, it helps to have a clear goal. You need to understand what the most important sounds are, how the song progresses and what it’s supposed to feel like. Is it meant to sound powerful, aggressive and big? Or is it more subtle, emotional and subdued? Should it be clean and commercial sounding, or more lo-fi and acoustic? Does it fit in with other songs of the same genre, or is it more experimental and unique?
That understanding comes from two considerations: context and references.
Start by listening to the song a couple of times to get a feel for it. What feelings and images does it create? Is it a song to dance to, a song that immerses you in deep listening, or more of a background track? It’s useful to hear it on a few different sound systems – try your studio monitors, headphones, earphones, even laptop and phone speakers. What elements stand out? Is it driven by lyrics, or an instrumental performance, or maybe the drums? This gives you a more real-world understanding, and a path toward the final mix.
If you’re working with a client, it’s helpful to ask as many questions as possible. What emotions or ideas are they aiming for? What inspired them? The more specific they are, the easier it’ll be to make decisions later down the line.
Once you understand the creative direction, it’s time to find a point of comparison. Gather reference tracks – existing songs that have similar ideas and sounds. These will be important for keeping you on track and guiding you towards the finished song. Tools like Cyanite.ai and MotionElements audio search can help find similar songs, but getting recommendations from clients will usually be more helpful for satisfying their expectations.
My method for using reference songs makes use of two plugins: MMultiAnalyzer and REFERENCE 2. These have different but equally useful tools for comparing your mix to the references, giving you information like the overall EQ curve of a song, the dynamic range and loudness levels, the stereo image and much more. All this information can be highly useful, especially when tweaking the finer details of a mix, but it shouldn’t replace a classic A/B test – your ears should guide most of your decisions.
The goal isn’t to copy the reference, but to learn from it. Figure out what works for your song, and forget about what doesn’t. You may find reference tracks useful for starting a mix, but as you get deeper into it the differences between your song and the reference will get more obvious, and eventually you won’t need to compare the two.
Once you have a clear vision for the song, it’s time to start building and organizing the mix.
2. Session Setup and Organization
In my experience, organizing the song is the most foundational (and the most overlooked) stage of mixing. It can be tempting to jump right in and start making changes, but if you skip this step you’ll really feel it later. Disorganised sessions can quickly become cluttered and confusing.
Start by organizing and labeling the tracks by sound type – I like to have the vocals grouped at the top, the melodics below, then the harmonies, bassline, percussion and finally the ambience, sound effects and transitional sounds at the bottom. Color coding them makes it much easier to visualize the different layers and labeling them will help when you’re making micro-adjustments and routing to buses.
This is optional, but I’m a big fan of heavy bus processing. All my main vocals get routed to a MV bus, backing vocals to BV bus, mid instruments to mid bus, percussion to drum bus and so on. This enables easier categorization, as well as giving you info about how the different sounds interact.
A key part of this is analysis tools – I use the plugin MMultiAnalyzer, which allows you to visualize any track in a number of different ways, particularly spectral and dynamic information. Here you can see how wide or narrow the drums are, how much room the vocals take up, and much more. My favorite feature is the ‘collisions’ section, which visualizes overlaps in frequency. It won’t solve the problem of your kick overlapping with the bass, or the guitar overtaking part of the vocal, but it’ll definitely show you it’s there, so that you can fix it later.
Another useful trick here is figuring out the fundamentals of each track – the base frequency that gives a track its pitch, usually the lowest frequency of significant volume. These fundamentals are often the most important frequency in a given sound, so it’s essential to make sure they’re not being masked by clashing frequencies from other sounds. For example, if the kick has a fundamental frequency of 55hz and the bass has one at 65hz, you could use an EQ to cut the frequencies in the bass that are lower than the fundamental, and another to dip the frequencies around 65hz in the kick. Using fundamentals to inform your decisions is key for clear and musical mixing.
This arrangement turns your project into a creative space that you know how to navigate. Now you’re ready to start shaping the sound.
3. Gain Staging and Technical Cleanup
Gain staging is the most common first change that a mixing engineer will make. It’s the process of making sure each sound is set at a healthy volume, which will create the initial balance between the elements in your track.
A time-tested approach is to start with all the volume faders at 0, and then bring the levels up to between -12 and -6dB, starting with the most important elements first. This is a good place to realize which parts should sit further in the background, and which parts should be upfront and clear. It’s important to make sure nothing is clipping (reaching 0dB), as this can affect the sound character and how it interacts with plugins. Leave lots of headroom (the gap between the loudest part of the song and the ceiling of 0dB) and your mastering engineer will thank you later!
Once the levels are balanced, you can move on to cleaning the tracks themselves. That can mean different things depending on the material. For anything recorded with multiple mics, start by checking the phase alignment. Phase alignment can be tricky, but the basic idea is that if you’ve recorded the same sound with two different mics, the tiny differences in timing can cause the sounds to cancel out and lose their power. Flip their polarity (most DAW samplers have this option) or use a phase alignment plugin like MAutoAlign to make sure this isn’t an issue.
From there, clean up any unwanted noise. This can be random sounds in the background of the recordings, reverb caused by echoey rooms, unwanted breaths in vocal recordings, the humming of electrical equipment or anything else that sounds out of place.
If you’re working with live recordings, using a high-pass EQ to cut out the low frequency noise that is unrelated to the actual sound (usually anything below the fundamental) is great for stopping low frequency ‘rumble’ from building up. Gate plugins are also great for this, but don’t overdo it, as they can also chop off parts of the sounds you want to keep. De-essers are useful for taming harsh sibilance – high frequency sounds like a vocalist emphasising the ‘s’ in a word. There are also many plugins for removing static or wind noise, room reflections and anything else unneeded, but be careful – too much editing can make a sound unnatural.
This stage is about cleaning, not enhancing or shaping the sound. With gain staging and noise removal, your track will already sound more controlled and professional, and will be ready for the next stage – creating space, clarity and separation in your sounds.
4. Carving Space — EQ, Ducking, Stereo Control
As mentioned, the goal of this stage is going beyond cleaning sounds, and creating separation. Separation is essential for clarity – making sure that every element is audible, so that nothing is fighting for space or being masked by other sounds.
First, use an EQ to make space in the frequency spectrum. This is where the analysis you did earlier will come in handy. Which elements should take up the lows, the low mids, the highs, and so on? Which elements are clashing? One way of approaching this is by making small reciprocal EQ moves – for example, you could boost the vocals at around 2-3kHz whilst gently cutting the guitar in the same range.
There is always the danger of mixing with your eyes instead of your ears, and it is important to make sure you’re hearing the difference made when EQing. Don’t just cut and boost based on what the analyzers are telling you, really try to figure out by ear where the clashes are and what sounds better to you.
Sometimes a fixed EQ isn’t enough – that’s when you can reach for a more reactive tool, like a dynamic EQ, multiband compressor or even a frequency specific sidechaining tool, like Oeksound’s Soothe 2. These make EQ changes based on input – you could use a multiband compressor on the high frequencies of a drum sample to keep the cymbals in check, a dynamic EQ to cut the lower frequencies of a piano track to avoid clashing with the bass, or use a spectral sidechain to duck the lowest frequencies in the bass out of the way of the kick drum. These time-based interactions are part of what gives a song a sense of movement and ‘breathing’.
Make sure to switch between listening to individual tracks and the whole song whilst doing this. There’s nothing worse than soloing a track, making it sound perfect on its own, and then finding out it no longer fits with the rest of the mix.
Once every track has its clear and defined space in the frequency spectrum, the next step is finding places for each element in the stereo image. This is more about perceived location than the actual sound character – it’s common to pan elements left or right to create separation, and to make them wide or narrow for the same reason.
For most engineers, the key sounds in a song are usually mono – right in the middle. These are typically the bass, kick, snare, the most important instruments and the main vocals. Having these elements in the center draws the listeners attention, and allows the supporting sounds to surround and complement them.
Also common is the idea that sounds should be wider the higher in frequency they are. Bass and kick sounds are almost always mono, whereas higher pitched pads, synths, backing vocals and cymbals are usually wider. This is a powerful effect that keeps the bass focused and grounded whilst the higher elements sound large and full. Remember, if everything is wide, nothing sounds wide. Contrast is what creates the feeling of depth.
After this stage, the mix should sound roughly ‘in-place’, without unwanted clutter, muddiness or masking. You’ve taken care of the frequency spectrum and the stereo image, but one important factor remains before you can get creative – the dynamics.
5. Dynamics and Energy — Compression, Clipping, Parallel Techniques
Controlling the dynamics of a track is like a secondary gain-staging – now you have everything at the right level, you can start controlling the loudness further, either to make it more consistent over time, or to emphasize certain parts. Your main dynamic tools are compressors, limiters and clippers.
Compression is the go-to tool for most dynamic processing. In essence, it sets a level of volume as the threshold, then reduces the volume of any sound that exceeds that threshold. This is key for maintaining a consistent level over the whole track, which prevents any sound from becoming too loud and drowning out the others. Any live recorded sound likely needs a bit of compression, especially key sounds that you want to keep consistent, like the bass, kick, snare, main instruments and vocals.
Use a moderate ratio (2:1 to 4:1) to gently control the levels, or a higher ratio (6:1+) to fix more significant volume issues. However, keep in mind that limiters and clippers are more designed for aggressive volume control, so you might want to use those for the more extreme cases, and to use subtle compression to get the overall balance right. A faster attack and slower release time will give you a more balanced overall sound, whilst a slower attack will allow the start of the sound to have more punch, and a faster release can create a pumping or breathing effect.
Limiters are similar to compressors, but they manage dynamics a lot more aggressively. Think of a limiter as a compressor with an extremely high ratio, like 100:1 – any sounds that exceed the threshold will be brought right down to the threshold level, instead of being reduced to be closer to it. Limiters aren’t as versatile as compressors, but can still be very helpful for sounds with huge dynamic ranges, or when dealing with multiple sounds at once in multi-track recordings.
Clipping has a similar function to limiting, but does it in a very different way. Instead of reacting to the audio when it exceeds the threshold, it simply cuts off the audio at the point. This takes a bit longer to understand, but it’s essentially the same as clipping accidentally when recording. If you’ve ever heard a recording that simply gets too loud for the microphone to handle and becomes distorted and squashed – that’s the extreme effect of clipping. However, more subtle uses, such as only clipping off 1 or 2dB, can control the dynamics in a much more direct way than something like compression. This is especially useful for keeping the punch of a very transient sound, like a kick or snare, whilst bringing the volume down so it doesn’t cover up the other sounds.
There are loads more dynamic tools, like transient shapers (which emphasize or soften the attack of sounds), expanders (which increase the dynamic range of sounds), inflators (which add perceived loudness without increasing the actual volume) and more, but compression, limiting and clipping will help you handle most situations. Now for the fun bit – adding color to your mix with saturation, distortion and creative EQing.
6. Saturation and Creative EQ
There’s only so far you can go with organizing, cleaning and controlling a song – at some point, you’ll want to make it pop. This is a much more free-form, genre-dependent and personal process, so it’s always best to find a method that works for you. That being said, I like to keep a track dry (without reverb, delay or other spatial effects) for as long as possible while I work on shaping the sound. This helps me dial in each track individually before they start coming together with shared reverbs, delays and bus processing.
Saturation is a key component of this process. Saturation, or distortion (they can be interchangeable, and for the sake of clarity I’ll refer to both as saturation), is a process created by clipping. The previous example of a recording that becomes too loud and clips is a good example for this too – the fuzzy, buzzing, overloaded sound that comes with the clipping is an extreme example of saturation. Similar to clipping, the usefulness of this comes from a much more subtle application. Saturation creates extra frequency information in a sound, making it sound brighter, larger and denser. You can use this to emphasize certain elements, to brighten up duller sounds, and even to add uniformity, by applying the same type of saturation to multiple sounds.
One reason for the popularity of saturation and saturating plugins is the way it emulates the analogue studio process. In a studio with hardware preamps, compressors, EQs, mixing desks etc, anything that runs through this analogue chain will be colored by the tiny imperfections in the circuitry, the cables, and anything else that is subject to damage over time. This adds tiny amounts of natural saturation at all stages, which adds up to create a harmonically complex and interesting sound. When people talk about the ‘warmth’ and ‘vibe’ of older music, this is most of what they mean.
For this reason, a lot of saturation plugins are meant to emulate the character of hardware. You can take this a step further by using multiple different instances of these plugins, with slightly different settings, to color a sound in multiple stages, just like an analogue workflow would. Or, if you want a more extreme result, grab a powerful saturation tool like FabFilter Saturn, and crank the settings all the way up. The best way to add creative effects is by hearing what the most extreme version sounds like, then dialing it back to be more tasteful.
Another key consideration at this stage is creative EQing. Earlier you used EQ to clean up and correct sounds – now it becomes a creative tool that you can use to emphasise or de-emphasise specific frequencies. This process is mostly about enhancing the character of sounds by bringing up key frequencies.
This could mean boosting the higher frequencies in a vocal (8kHz – 14kHz) to clarify the pronunciation of the lyrics, adding punch to a snare by boosting the lower frequencies or the fundamental, bringing up the tone of a guitar in the high-mids (2kHz – 4kHz) or enhancing the ‘air’ by boosting sounds above 14kHz to 16kHz.
It’s worth being cautious here – boosts and creative EQ decisions should be made to enhance the sound, not to force it into a space that it doesn’t naturally exist in. As always, you should be making decisions in the context of the mix, instead of soloing tracks and trying to make the best individual sounds.
7. Depth, Space and Movement — Reverb, Delay, Automation
Now you’ve worked on the balance and energy of the mix, it’s time to add movement. This is where the song comes to life, moving from a static sequence of sounds to a cohesive combination that ‘breathes’.
Reverb is the most direct way to add depth. It tells the listener ‘where’ the song is, whether it’s a tight, controlled room, a concert hall, or an expansive dreamscape. By creating reverb send tracks, you can standardize this sense of space. To do this, insert a reverb plugin like FabFilter Pro-R or your DAWs stock EQ on an empty track, and turn the mix knob to 100% wet. Now you can route other tracks through this ‘send’ track, and it’ll only play the sound of the reverb, keeping the dry signal separate.
Shorter, more subtle reverbs can be used to put a sound (or a group of sounds) in a ‘room’. By giving multiple sounds a short reverb, you can make them sound as if they were recorded in the same space, gluing them together and enhancing the connections between them.
Longer reverbs push a sound towards the background, which works well on backing vocals, sustained instruments, pads and any effects. It’s also important to keep some sounds dry – if there’s reverb on everything, the sounds will start to blur together and sound muddy and disorganized.
Delay is another key tool for movement and depth. It’s easy to think of it as just an echo effect, but it can do so much more. For example, a very short stereo delay (10 – 30ms) can be panned left and right to create a sense of width. Medium length delays (eighth or sixteenth notes) can create an internal bounce and rhythm. Longer delays can be used as transitions where a sound continues echoing out after it stops playing or can be used as rhythmic devices on their own.
As with reverbs, delays can be used on send tracks for cohesion between sounds. Both ‘sends can benefit from EQing, as too much movement in the lowest and highest frequencies can be distracting for the listener. Most successful uses of both are calculated and subtle, creating depth and movement that is felt rather than heard.
Automation is a much broader category. It essentially means changing any parameter over time, whether it’s automating the volume of a key instrument to add more energy to certain sections, automating the delay to make a transition pop, or automating EQ to slowly introduce a sound. Automation is a key step for the emotion, tension and evolution of a song.
There are no strict rules for this, so it’s best to experiment, focusing on the key sounds for maximum impact and making smaller changes to the background sounds to keep the song alive. It’s definitely worth spending a decent amount of time on this section.
8. Final Touches — Bus Processing and Print-Ready Polish
Now the track is clean, controlled, expressive and moving, with the beginnings of cohesion and uniformity. The final step in this guide is bus processing – moving from processing individual tracks to working on groups of tracks. As the most macro step yet, subtlety is extremely important. Very deliberate, gentle changes, and a lot of A/B testing is advised.
Start with the main buses. These will vary depending on the song, but a popular technique is the Bruaerizing setup, which includes bass, percussion, vocals, mid-range instruments and background elements. Your bus choices will depend on the artist, genre, and your overall mixing philosophy – I use a lot more buses and sends than is common. Once these buses are set up, route every track of that category to its bus.
Most bus processing revolves around glue compression (using subtle compression to ‘glue’ sounds together by giving them more uniform dynamics) and saturation. These are the key tools for making the sounds cohesive, which is the main goal of this stage. There are also reasons to use additional transient shaping, stereo shaping, clipping and creative EQ – it’s all about what best serves the song.
It’s important to remember that any changes you make to buses will have a strong effect on the track as a whole. Most of the more dramatic changes should have already happened by this point, and bus processing should be a subtle and controlled process.
The final step should be to check all the measurable parameters. This is usually the peak loudness, average loudness, overall EQ curve and the stereo image. You could use this as a final chance to go back to your original reference tracks and see how closely your mix compares, once again making use of metering tools like MMultiAnalyzer and REFERENCE 2. Your mix should peak at around -6dB, and should have strong coverage across the frequency spectrum and stereo space.
This is also a good time to check the mix on different playback systems. As with the first listen, a quick check on studio monitors, headphones, earphones, laptop and phone speakers may reveal an issue with the balance or give you a new perspective on the track.
Conclusion
Mixing is a craft that engineers hone over decades, so don’t be discouraged if it takes a long time when you’re first starting out. It’s not just a routine step in the song making process, but another opportunity for creativity and enhancement, as much an expressive art as a scientific process. Building a system is a key step for beginners, to remind yourself of the ideas to consider and the tools you have available. Personalizing your system will help you mix intentionally and understand what a song needs to take it to the next level.
Whether mixing your own song or working to client expectations, the mixing process is a chance to test your skills and express yourself. Don’t get too bogged down with technicalities, don’t be afraid to experiment, and most importantly always trust your ears!
For a limited time, the expert engineers at We Rave You are offering a free mix and master review service. Drop us a mail at hemant [at] weraveyou.com, and get your mix or master reviewed by our experts.
RELATED: Mixing Music on Headphones: Pro Tips for Killer Mixes Without Monitors
Also Read:
