prep your files.

• Wet multitracks — the main set. Your processing printed, nothing on the master/mixbus. This is what I mix from — I start where you left off, not from zero.
• Dry multitracks. The same tracks, unprocessed — identical file names, same start point. When I hear more in a part — the vocal, most often — the dry take lets me build the sound up from scratch instead of stacking my processing on top of yours.
• Your project, frozen. The full DAW session with every track frozen or committed to audio — that's what makes it open cleanly on my system, no missing plugins, no crashes.
• Rough mix — plus references, if you have them. The bounce you're happiest with is a must. References are optional, and no files needed — artist + song in a text file, or a link to a Spotify/streaming playlist, is plenty. What helps most is why you picked them: "the vocal on this one, the low end on that one" beats "make it like this song".

Any DAW works — Logic, Ableton, FL, Pro Tools, Studio One, Cubase… The reason I ask for multitracks instead of your session: I mix in my own system, the one I know inside out — and that's what gets your record the best-sounding result.

• Finish the editing first. Tuning, timing, comping and arrangement final — and print the Melodyne/Auto-Tune so it's baked into the audio. I mix from where you left off; I'm not re-editing the song.
• Keep your sounds. If an effect is the part — a printed delay, a crushed drum bus — it belongs in the wet set. The dry set is the safety copy; don't sanitise the wet one.
• Group buses work like the mixbus. Glue on a group — say, your drum bus — won't print when you export individual tracks, so they'd come back sounding different from your rough. Same routine: screenshot the chain into REFERENCE, then bypass it. Unsure about your case? Just ask.

Format

WAV

Bit depth

32-bit (float)

Sample rate

Same as your session — don't convert it

Start point

All files from bar 1 / 0:00, same length

Mixbus

Bypass your master/mixbus chain while exporting — but screenshot it first (plugins, settings, order) and drop the image in the REFERENCE folder

Also include

Tempo / BPM & key — they go in the folder name

Same start point on every file is the big one — everything lines up instantly on my end. And 32-bit float can't clip on export, so levels are one less thing to babysit.

Don't build this by hand — the template above has it all, with empty example files showing the naming. Rename the main folder to your song:

The file numbers are my session order. Sorted A–Z, your files build my session automatically: vocals at the top, bass right above the drums so the low end stays together, then music and FX. Keep the numbers, change the words.

Good	Not this
01 Lead Vox	Audio 14
10 BGV Stack L	harms comp 2
30 Bass DI	untitled bass
45 Snare Top	SnrTopFINAL_v3
70 Gtr Rhythm L	gtr bounce (use this one)

01–09 lead vocals · 10s backing vocals · 20s vocal FX · 30s bass · 40–69 drums (40 kick, 45 snare, 50 toms & low perc, 60 hats & cymbals) · 70–89 music (70 guitars, 80 keys, 85 other) · 90s FX. The dry set uses the exact same names as the wet set — that's how the two pair up on my end. Folders that don't apply can stay empty.

Multitracks are the individual tracks in your session. If your kick is three layers — kick in, kick out, a sample — that's three files. Stems are groups summed into one file: those three kicks bounced together as a single "kick stem".

Multitracks are my preference. Separate layers let me rebalance them, fix the phase between mics and samples, or ease a layer back when the mix wants less of it. A stem locks those choices in — which is fine if you love the blend and want it kept.

One exception: if a section is huge — an orchestra, a wall of layered sound design — and it's already edited and balanced exactly how you want it, bounce it to a stereo stem. Better that than me wading through eighty tracks you've already perfected.

I'll happily work from either — just know what each one means: multitracks give me more ways to make it better; a stem tells me "this part is done".

Three versions sounds like overkill. Here's what each one buys your mix:

• Wet is your sound. Processing printed, decisions intact. It's what I mix from, so nothing you loved gets lost on the way to me.
• Dry is room to improve. When I hear more in a part — the vocal, most often — I rebuild the processing from scratch instead of stacking mine on top of yours. That only works from the dry take.
• The frozen project is the safety net. Every track committed to audio, so your session opens on my system without your plugins. If the stems ever leave a question — an odd noise, "which file is the real lead?" — I check the source instead of holding up your mix.

Ninety percent of the mix happens on the wet set. The other two are what let me chase the last ten percent without bothering you.