The short answer: Send clearly labeled stems or a session exported at the same start point, all at the original sample rate and 24-bit resolution, with nothing clipping, no mix-bus limiter printed in, a reference track or two, your tempo and key, and a short note about what you want. Do that and you save yourself a round of revisions before the work even begins. Get it wrong and even a great engineer is fighting your prep instead of shaping your song.

Most "why doesn't my mix sound pro" problems trace back to what showed up in the inbox, not to the plugins that came after. Here is exactly how to hand off a session that gets you the best possible result.

Get the files right first

This is the part people rush, and it costs them the most. Nail the technical layer and everything downstream gets easier.

  • Send WAV or AIFF, never MP3. MP3 throws away detail permanently. If your only copy is an MP3, say so up front so nobody assumes it's a full-quality file.
  • Match the project's sample rate and bit depth. Export at whatever the session was recorded in, typically 44.1 or 48 kHz, and at 24-bit. Don't upsample or convert on the way out; it adds nothing and can introduce artifacts.
  • Export every track from the same start point (bar 1, time zero). This is the single most important rule. When all stems begin at the same timestamp, they drop straight into a session perfectly aligned. Consolidate each track from the top, even if it stays silent for a while.
  • Leave headroom. Aim for peaks around -6 dBFS and nothing touching 0. A file that's clipping can't be un-clipped. Turn things down before export, not after.
  • Bounce dry unless an effect is part of the sound. Print your creative delays, the specific reverb on a vocal ad-lib, the distorted bass tone you built the song around. Skip general "make it sound finished" processing so the engineer has room to work.

Stems vs. a full session

There's no universal right answer here; it depends on how you work and who you're sending to.

  1. Consolidated stems are the safest, most portable option. One audio file per track or per group (drums, bass, lead vocal, backing vocals, keys, etc.), all starting at zero. These open in any DAW and are the standard for remote mixing.
  2. A full DAW session works only when you and the engineer use the same software and you've confirmed they have your plugins. Otherwise, stems are the move.
  3. Group your stems sensibly. Keep the lead vocal on its own. Keep the kick and snare separate if you want real control over the low end and punch. Don't collapse a 40-track session into three stems and expect surgical results.

Label everything plainly: "Lead Vox," "BGV Stack," "808," "Kick," not "Audio 7" or "Track 12 final FINAL." Clear names save real time.

The myth worth killing: "The engineer will fix it in the mix." Mixing is balancing and shaping what exists, not rebuilding it. A muddy recording, a vocal drowning in room noise, a badly tuned take, or timing that's all over the place stays a problem after mixing. Fix the source first. The best mixes start with the best recordings, and no plugin substitutes for a good take in a treated space.

The context that makes a mix land faster

Files tell the engineer what you have. Context tells them where you're headed, and that's what turns a technically clean mix into the one you actually wanted.

  • One or two reference tracks. Commercial songs whose sound you're chasing, whether it's the vocal presence, the low-end weight, or the overall vibe. This communicates more in thirty seconds than a paragraph of adjectives.
  • Tempo (BPM) and key. Especially if there's any tuning, time-based effects, or additional production involved. Include it in your notes even when it's in the session.
  • A rough mix. Your own balance, however imperfect. It shows the engineer your intent, the moments you care about, how loud you hear the vocal sitting. They don't have to follow it, but it's a map.
  • A short, specific note. Three or four lines. "Vocal forward and intimate, 808 heavy but clean, keep it dark, the drop at 1:45 should hit hard." Vague briefs like "make it sound professional" force guesswork and extra revisions.
  • A lyric or arrangement sheet if the vocal has fast sections, harmonies, or ad-libs that are easy to miss.

A pre-send checklist

Run this every time before you hit send. It takes five minutes and prevents the most common do-overs.

  1. Every stem exported from bar 1 / time zero.
  2. All files WAV, 24-bit, at the session's native sample rate.
  3. No track clipping; peaks with comfortable headroom.
  4. No limiter or heavy processing on the master/mix bus.
  5. Tracks clearly named and grouped.
  6. Reference track(s) included.
  7. Tempo and key noted.
  8. Rough mix included.
  9. A short written note on the direction you want.
  10. Everything zipped into one clearly named folder, sent via a link that keeps full quality (a file-transfer service, not a compressed email attachment or a messaging app that recompresses audio).

One more thing on delivery: name the folder with the song title and version, like "Midnight_v3_stems." When revisions happen, and they will, clear version names keep everyone working on the same file instead of an old one.

When you want it to sound truly professional

Great prep gets you most of the way, but there's a point where the song deserves an experienced ear, someone who's mixed the genre you're working in and knows how to make a vocal sit, a low end translate on phone speakers and in the car, and a record feel finished rather than just loud.

That's the work I do every day with independent artists and small labels. If you're sitting on a song you believe in and you want it to compete, send it over. Every project is different, so pricing and turnaround are handled case by case once I hear what you're working with and what you're aiming for.

Head to the contact page, tell me about the record, and book a call. Bring your rough mix and a reference or two, and we'll talk through exactly what your song needs to sound like the version you hear in your head.