How to Prepare and Send Stems to a Mixing Engineer
You’ve found your mixing engineer, agreed a price and booked a date. The next thing that decides how good the mix will be isn’t their plugins. It’s what you send them.
A clean stem pack means the engineer spends their first hour mixing. A messy one means they spend it renaming files, asking which of the three vocal comps is the real one, and waiting for you to re-export the bass that came out half a bar late. Every one of those questions costs a day.
This guide covers how to export stems so they arrive ready to mix. If you’re still looking for someone, start with our guide to hiring a mixing engineer online.
Stems vs. Multitracks: What Are You Sending?
The word “stems” gets used for two different things:
- Multitracks (individual tracks): every track in your session exported separately. Kick, snare, overheads, bass, each guitar, each vocal. This is what most mixing engineers want.
- Stems (grouped tracks): submixes of related tracks, like all drums on one file and all backing vocals on another. These are common for mastering, live playback or remixes, but they limit what a mixing engineer can do.
When an engineer says “send me the stems”, they almost always mean individual tracks. If you’re not sure, ask. It’s a 30-second question that can save a re-export.
Before You Export: Clean Up the Session
Five minutes of tidying makes the export faster and the files easier to understand:
- Commit to your takes. Comp the vocals, pick the final guitar take, and delete or mute everything you’re not using. Don’t send three versions of a part and expect the engineer to choose.
- Render virtual instruments and MIDI to audio. The engineer may not have your plugins.
- Name your tracks in the DAW before exporting, because most DAWs use track names for file names.
- Don’t trim silence. Keep every track full length, even if a part only plays once (see below).
How to Export: The Settings That Matter
Start every file at the same point
Export every track from the same start point, usually bar 1 or the very start of the session, and make every file the same length, even if the instrument only plays in the last chorus. When the engineer drops the files into their session, they line up automatically.
This is the single most common problem with stem packs. A backing vocal that starts where the singer starts, instead of at bar 1, has to be placed by ear.
Keep your sample rate and bit depth
Export at the same sample rate as your session (44.1 kHz, 48 kHz, 96 kHz). Don’t convert it. For bit depth, use 24-bit or 32-bit float WAV, whichever your engineer prefers, and don’t export at 16-bit. 32-bit float needs no dither; if your DAW offers it when you export processed tracks at 24-bit, it does no harm, but never dither twice.
WAV or AIFF are the safe choices. Never send MP3s as stems.
Leave headroom, and turn off the master bus
- Turn off anything on your master bus: limiters, compressors, EQ, “mix bus chains”. Those belong to the mix, and baking them into stems ties the engineer’s hands.
- Leave headroom. No file should clip. Peaks somewhere around -6 dBFS give the engineer room to work. If a track is clipping, pull its fader down before exporting; don’t normalize.
Decide what to do with effects
A good rule:
- Export dry (no reverb, delay or bus processing) by default.
- If an effect is part of the sound (the tape delay on the guitar line, the amp sim that makes the bass, a creative vocal effect), export it with the effect and send a dry version too, labelled clearly.
- Leave out your mix decisions: reverbs on sends, parallel compression, bus glue. The engineer will make their own.
If you’re attached to a reverb or delay sound, print it as a separate wet-only file and say why in your notes.
Mono or stereo?
Export mono sources (a single kick mic, a DI bass, a lead vocal) as mono files, and stereo sources (overheads, a stereo synth, a stereo room) as stereo. Exporting everything as stereo doubles file size and makes the engineer check which files are fake stereo. If your DAW only exports stereo, say so in your notes.
Naming and Organizing the Files
Name every file so someone who has never heard the song knows what it is:
01_Kick_In.wav
02_Kick_Out.wav
03_Snare_Top.wav
04_Snare_Bottom.wav
10_Bass_DI.wav
11_Bass_Amp.wav
20_Gtr_Rhythm_L.wav
21_Gtr_Rhythm_R.wav
30_Synth_Pad.wav
40_Vox_Lead.wav
41_Vox_Double.wav
42_Vox_Harmony_High.wav- Number them so they load in a sensible order: drums, bass, guitars, keys, vocals.
- Avoid vague names:
Audio 14.wav,New Track.wavandvox final FINAL.wavall mean extra questions. - Put the song title on the folder, not on every file name, unless you’re sending several songs at once.
What Else to Include
The audio isn’t the whole delivery. Add:
- A rough mix of where the song is now. It tells the engineer what you’ve been hearing.
- Reference tracks, two or three commercial songs with a sound you’re aiming for, and a sentence on what you like about each.
- A short notes file with the tempo (BPM), key, sample rate, anything unusual (“Gtr_Lead has amp sim printed; dry version is Gtr_Lead_DI”), and anything you want kept or changed.
- A mix brief: what the song should feel like, and any non-negotiables. The hiring guide has a template.
Sending the Files
Stem packs are big. A song with 30–60 tracks at 24-bit often runs to a gigabyte or more, so email is out. Zip the folder so nothing gets lost, and send it with a link that won’t expire before the engineer downloads it.
Once you’ve sent it, ask the engineer to confirm everything opened and lined up. That’s the moment to catch a missing file, not halfway through the mix.
The Pre-Send Checklist
Run through this before you hit send:
- Every track exported, including the quiet ones you forgot about
- All files start at the same point and are the same length
- Session sample rate kept; 24-bit or 32-bit float WAV; no MP3s
- Master bus processing turned off; nothing clipping
- Effects that are part of the sound printed, plus a dry version
- Mono sources exported as mono
- Files numbered and named clearly
- Rough mix, references and notes (BPM, key, sample rate) included
- Folder zipped and sent with a link that won’t expire soon
It looks like a lot, but once it’s a habit it takes ten minutes, and it’s the difference between a mix that starts today and one that starts after three rounds of “can you resend the bass?”
After the Handoff
Once the mix is underway, the same discipline carries on: keep every mix version numbered, put feedback against a specific version, and answer every note. Our posts on why feedback workflows break and audio file version control cover that part.
If you’re on the other side of the handoff, our guide to organizing audio files has a checklist for engineers receiving a stem pack.
Sharing Stems With Aliada
With Aliada, the stem pack and every mix version live in the same project. You upload the stems once (as individual WAV or FLAC files rather than a zip), your engineer uploads each mix as a new version of the track, and you leave timestamped notes on the waveform of the version you’re hearing. Every version keeps its untouched lossless original, so nobody has to dig through old links to find last week’s mix.
Try Aliada free on your next session.