Every mix engineer knows the feeling of opening a session and sensing trouble before touching a single fader. Not because the performances are weak, but because the recordings themselves have made the job harder. Noise floors sitting high. Tracks recorded so hot they clip. Thirty audio files named “Audio 01,” “Audio 02,” and “Audio 02 FINAL real.” None of these are mixing problems. They are recording problems wearing a mixing costume.
The uncomfortable truth is that most mixing headaches are decided in the room, long before the mix begins. Gain, mic placement, arrangement density, monitoring, file management: these choices shape the large majority of what a mix engineer ends up working with. The craft of mixing handles the rest. So the single most valuable habit a less experienced engineer can build is treating the session itself as where the mix is won or lost, which starts with preparing properly before anyone hits record. Here are the mistakes that turn up most often, and what to do instead.
Recording too hot
This one is a holdover from a different era. On tape and in early digital, you chased the loudest clean signal you could get, because the noise floor was real and audible. In 24-bit recording that logic no longer holds. The noise floor is so far down that conservative levels cost you nothing, while hot levels risk clipping the converter, the one mistake you genuinely cannot undo later.
Aim for peaks somewhere around -18 to -12 dBFS, with plenty of room above. It will look quiet on the meter. It will sound perfect, and it leaves the mix engineer a clean, undistorted signal to push however they like. A track that arrives quiet is a non-issue. A track that arrives clipped is a problem with no good solution.
Ignoring phase
Any time a single source is captured by more than one microphone, phase matters. Snare top and bottom. A guitar cabinet with a close mic and a room mic. A DI and an amp running together. When those signals reach the converter slightly out of time with each other, they cancel, and the result is a thin, hollow sound that no amount of EQ will properly rescue.
The fix sounds technical but is simple to hear. Before you commit, flip the polarity switch on one of the paired mics and listen to the low end. If the bottom suddenly thickens when you flip it, or drops out, you have your answer about which position holds together. You are not looking for a number on a screen. You are listening for the version where the source sounds full and solid, then leaving it there.
Recording everything, then sorting it out later
A dense arrangement feels productive in the moment. Six guitar layers, three tambourine passes, a synth pad doubling the piano doubling the strings. Every part justified on its own. Stacked together, they fight for the same space, and the mix engineer inherits the job of deciding what survives, which is really an arrangement decision made too late.
Make those calls in the room while the artist is present and the creative context is fresh. If a part is not earning its place, do not print it. The fewer competing elements that arrive, the more room there is to make the ones that matter sound genuinely good.
Forgetting that monitoring shapes the performance
Handing over a session with no map
Organisation sounds dull right up until it becomes the difference between a mix that starts in ninety minutes and one that starts in four hours, after the engineer has reverse-engineered what every track is and where it belongs.
Name tracks by instrument and function. Colour-code the session. Drop markers at the song sections. Consolidate alternative takes instead of scattering them down the timeline. Leave a short notes document covering the artist’s intent, the reference tracks, and any technical decisions made on the day. These habits cost nothing during the session and pay back completely the moment someone else opens the file.
Nobody ever walked out of a great mix session wishing the recordings had arrived sloppier. The room is where it starts. Everything after that is just consequences.
Article provided by
