How to Avoid File Format Mistakes When Ordering Mastering
Avoid mastering delays by sending one approved stereo premaster as a lossless WAV or AIFF at the session's native sample rate and bit depth, with no clipping, no lossy conversion, no accidental sample-rate change, no surprise limiter, and a clear filename. Mastering starts faster when the engineer receives the exact mix you approved, not an MP3, not a normalized bounce, and not a folder of mystery versions.
Send the right file once so the master can focus on translation, loudness, and release quality.
Book Mastering ServicesFile-format mistakes are one of the most avoidable reasons mastering gets delayed. The song may be mixed well, the artist may be ready, and the release date may be close, but a bad export forces the engineer to stop and ask for a new file. That back-and-forth costs time and can create confusion about which version is final.
Most of these mistakes happen in the last five minutes of the mix session, when everyone is tired and trying to send the file quickly. Slow down before export. The bounce settings are part of the record now, not an administrative detail.
The right mastering file is not complicated. It is the clean, approved, full-quality stereo mix before final mastering. Most of the time, that means a WAV or AIFF file exported from the mix session at the same sample rate and bit depth as the project. It should have enough headroom to work with, no clipping, no unintended limiting, and no file conversion that throws away audio detail.
That clean handoff protects the song, the deadline, and the budget. No guessing. No confusion.
If you are ordering mastering services, the export is part of the job. A mastering engineer can shape tone, loudness, sequencing, spacing, translation, and final quality control, but the engineer should not have to guess which bounce is real or rebuild quality that was lost in an MP3. The better the file you send, the cleaner the mastering decision becomes.
The Mastering File Is Not the Release File
One common confusion is mixing up the premaster with the final release master. The premaster is what you send to mastering. The release master is what comes back after mastering. Do not send a streaming-normalized file, an MP3 preview, a clipped loud reference, or the file your distributor will eventually receive unless that is the actual unmastered mix at full quality.
A premaster should preserve the mix session's quality. The mastering engineer then creates the final master for release needs. Apple Digital Masters guidance, for example, discusses high-resolution source files and native project resolution, but that does not mean you should fake a higher resolution export after the fact. If the session was recorded and mixed at one quality, preserve that quality instead of upsampling or converting just to make the numbers look bigger.
| File type | Use for mastering? | Why |
|---|---|---|
| WAV | Yes | Lossless, common, full-quality delivery format |
| AIFF | Yes | Lossless and acceptable in many professional workflows |
| MP3 | No, unless there is no other option | Lossy compression throws away audio information |
| AAC or M4A | No | Designed for delivery or listening, not mastering intake |
| FLAC | Sometimes, if accepted | Lossless, but not always the requested intake format |
| Stems | Only when requested | Stem mastering is a different workflow from stereo mastering |
If you are deciding whether stems are needed, use the framework in stereo mastering vs stem mastering. Do not send stems because you are unsure. Send the stereo mix unless the mastering plan specifically calls for grouped stems.
Mistake 1: Sending an MP3 or Other Lossy File
MP3 and AAC files are useful for texting a preview, checking a song in the car, or sending a quick reference. They are not ideal mastering sources. Lossy formats reduce file size by removing audio information. That damage may be subtle when listening casually, but mastering can reveal artifacts, smear transients, and make high-frequency problems more obvious.
LANDR's upload guidance recommends avoiding lossy conversion and uploading the rawest available mix. That is the right principle even when you are not using automated mastering. The mastering source should be the most complete version of the mix, not a compressed copy of it.
If the only file you have is an MP3, be honest with the engineer before ordering. It may still be possible to improve it, but the expectations should be different. A master from a lossy file cannot regain the original recording detail that was removed during encoding.
Mistake 2: Changing Sample Rate for No Reason
The safest export is usually the session's native sample rate. If the project was recorded and mixed at 48 kHz, export the premaster at 48 kHz. If it was built at 44.1 kHz, export at 44.1 kHz. Do not convert to a different sample rate just because a random forum said another number is better.
Unnecessary sample-rate conversion adds another process that can create confusion or artifacts. The mastering engineer can handle final delivery versions after the master is approved. Your job at the premaster stage is to preserve the actual mix as cleanly as possible.
Avid's Pro Tools bounce workflow shows that file type, stereo format, bit depth, and sample rate are explicit export choices. That means it is easy to choose the wrong setting if you rush. Before bouncing, check the session settings and confirm the export window matches the intended delivery.
Mistake 3: Exporting the Wrong Bit Depth
For most mastering orders, 24-bit WAV or AIFF is a safe, professional choice if the session supports it. Some DAWs can export 32-bit float, and some engineers accept it, but 24-bit is widely understood and preserves plenty of dynamic detail. The mistake is not only choosing 16-bit. The bigger mistake is changing the bit depth without knowing why.
Do not export a 16-bit file because it seems more like a final consumer file. That is a release-delivery mindset, not a mastering-intake mindset. Also avoid fake quality moves. If the session was not actually captured at a higher quality, converting after the fact does not create new detail. Preserve the real mix quality and ask the engineer if a specific format is needed.
Mistake 4: Normalizing the Bounce
Normalization raises or adjusts level based on the file's peak. That can make a file look tidy, but it is usually not what a mastering engineer needs from a premaster. If the mix already has healthy headroom, normalization changes the level relationship for no musical reason. If the mix is clipped, normalization does not repair the clipping.
Leave the level natural. A common safe target is peaks around -6 dBFS, but the exact number is less important than avoiding clipping and leaving room for mastering moves. LANDR's premaster guidance puts the main concern clearly: do not exceed the digital ceiling. If your meter hits red, fix the mix or master-bus processing before export.
Do not chase headroom by lowering the final file after it already clipped. If the master bus clipped during bounce, pulling the file down afterward only makes a quieter clipped file. Lower the mix bus, limiter input, or track levels before exporting again.
Mistake 5: Leaving a Limiter on the Master Bus by Accident
A limiter can be part of your mix if you deliberately mixed through it and the sound depends on it. But an accidental loudness limiter on the master bus is a problem. It can reduce transient punch, flatten the low end, create distortion, and leave the mastering engineer with less room to work.
Before exporting, check the master bus. If there is a limiter, clipper, maximizer, loudness chain, or analyzer with gain engaged, decide whether it is part of the approved mix or only a temporary loudness preview. If it is only for previewing, bypass it for the premaster and send a separate loud reference if useful.
This is also a good time to compare the premaster against the loud reference. The unmastered file should be quieter. That is normal. Do not panic and turn it up because it does not compete with released songs yet. Mastering handles final loudness after tone and translation are right.
Mistake 6: Sending the Wrong Version
Version confusion can waste more time than technical settings. A folder with "final," "final2," "real final," "artist notes," "new hook," and "loud bounce" invites mistakes. The engineer should not have to guess which file is approved. The artist should not discover after mastering that an old vocal edit was used.
Use clear filenames. Include artist, song title, mix version, and date or version number if needed. Avoid vague names like "bounce.wav." If there are clean and explicit versions, label them clearly. If there are alternate endings, label them clearly. If there is one final mix, send one file and say that it is the approved premaster.
Before uploading, play the entire file from start to finish outside the DAW. Listen for missing intro, cut-off tail, wrong vocal, muted ad-lib, broken automation, or accidental solo/mute states. Many export mistakes are not format mistakes; they are attention mistakes.
Mistake 7: Sending Stems When Stereo Mastering Was Ordered
Stem mastering uses grouped stereo files like drums, bass, music, lead vocal, and backgrounds. It can be useful when a mix has a few balance issues that cannot be fixed by stereo mastering alone. But stems create more responsibility. They must line up perfectly, sum to the approved mix, share the same start point, and include the right processing choices.
If you ordered stereo mastering, send the stereo premaster. If you think the mix needs stem mastering, discuss that first. Stems sent without context can slow the job down because the engineer has to ask whether they are supposed to master the stereo file, rebuild the mix from stems, or use stems only for problem solving.
DAW Export Settings That Usually Work
Every DAW labels export options a little differently, but the practical goal is the same: capture the stereo mix bus exactly as intended at full quality. In Pro Tools, Avid's bounce workflow shows options for file type, channel format, bit depth, sample rate, and bounce behavior. Logic, Ableton, FL Studio, Studio One, Reaper, Cubase, and other DAWs expose similar choices even if the names change.
For normal stereo mastering, the export should usually be stereo interleaved, not split mono. It should cover the full song from the true start to the end of the last reverb or delay tail. It should use the main stereo output or mix bus that you actually approved. If your DAW has options like "normalize," "convert to MP3," "add to library," or "render as loop," leave them off unless your engineer specifically asks for that behavior.
| Export setting | Safe mastering choice | Common mistake |
|---|---|---|
| File type | WAV or AIFF | Choosing MP3 because it uploads faster |
| Channel format | Stereo interleaved | Sending left/right split files by accident |
| Sample rate | Native session rate | Converting because another number looks better |
| Bit depth | Usually 24-bit, or engineer-requested format | Choosing 16-bit because it feels like a final product |
| Normalize | Off | Letting the DAW change level after the mix is approved |
| Range | Full song plus tails | Cutting the first transient or ending reverb early |
If your export window includes dithering, ask whether it is needed. Dither is normally used when reducing bit depth, not as a random quality enhancer. If you are exporting a 24-bit premaster from a 24-bit workflow, you may not need it. If you are not sure, leave a note for the engineer instead of guessing.
Reference Files Are Helpful, But Label Them Clearly
It is fine to send a loud rough master, a previous demo, or a commercial reference. The mistake is making those files look like the actual premaster. Put references in a separate folder or label them clearly as references. The mastering engineer should never wonder whether "song_loud.wav" is the file to master or only the version you used for listening in the car.
A loud rough can be useful because it tells the engineer what the artist has been hearing. It can also show whether the mix depends on a limiter's tone. But the rough should not replace the clean premaster. Send both only when it helps: one file marked "premaster" and one file marked "loud reference." That naming habit prevents avoidable mistakes when several people are involved.
Red Flags Before You Upload
Stop and rebounce if you notice any of these issues. A mastering engineer can catch some of them, but you should not rely on mastering intake to find basic export problems.
- The file is much smaller than expected for a WAV or AIFF.
- The bounce opens as mono even though the mix is stereo.
- The beginning cuts off the first kick, breath, or pickup.
- The ending cuts off reverb, delay, or a fade.
- The vocal version is not the approved take.
- The master bus meter clipped during export.
- The file sounds louder than the approved mix because normalize was on.
- The limiter was left on even though it was only for preview loudness.
- The filename does not identify the artist, song, or version.
- You cannot explain what each file in the upload folder is for.
The last point is the real test. If the folder is confusing to you, it will be confusing to the engineer. Clean the upload before sending it.
Pre-Upload Mastering Checklist
Use this checklist before you upload the file:
- The mix is approved by the artist and producer.
- The file is WAV or AIFF unless the engineer requested another lossless format.
- The sample rate matches the session's native sample rate.
- The bit depth is appropriate for mastering, usually 24-bit if available.
- The master bus is not clipping.
- Peaks leave safe headroom and do not hit 0 dBFS.
- Preview-only limiters, clippers, and loudness maximizers are bypassed.
- Creative mix-bus processing that defines the mix is intentionally left on.
- The file starts and ends correctly, including reverb and delay tails.
- The filename clearly identifies artist, title, and version.
- The whole bounce has been played outside the DAW.
- Reference notes are included separately, not baked into the file name.
This is the same kind of discipline behind a strong pre-mastering checklist. Mastering is not where you discover that the second verse was muted or that the mix was exported from the wrong session copy.
What to Send With the File
The audio file matters most, but context helps. Send the artist name, song title, preferred version name, references, release goal, and any known concerns. If you want the master to stay dynamic, say that. If the song needs to compete in a loud genre, say that. If the mix has a low-end issue you are worried about, say that. Clear notes help the engineer make better decisions.
Keep the notes practical. "Make it sound amazing" is not useful. "The reference has a tighter kick and a brighter vocal, but I do not want the hi-hats to get sharper" is useful. "The client likes the loudness of the rough limiter, but the unmastered premaster has the limiter bypassed" is useful. The goal is to give direction without asking mastering to become mixing.
After the master comes back, judge it against the right standard. Do not only ask whether it is louder. Use the mindset from spotting a mastering demo that is just louder, not better. A good master should translate, feel balanced, preserve emotion, and improve release readiness.
When to Go Back to Mixing Instead
Sometimes the mastering file is technically correct but the song is not ready. If the vocal is too quiet, the 808 is distorting, the snare is painfully sharp, or the hook balance is wrong, a clean WAV will not solve the creative issue. Mastering can polish and correct, but it should not be expected to rebuild the mix.
Use the DIY mastering checklist as a final self-check. If the mix fails a basic translation, clipping, mono, vocal-balance, or low-end test, fix the mix first. The best mastering result comes from a mix that already feels like the record, just not finished for release yet.
FAQ
What file format should I send for mastering?
Send a lossless WAV or AIFF file unless your mastering engineer asks for something else. Use the native sample rate and bit depth from the mix session, with no clipping, no lossy conversion, and no accidental loudness processing.
Can I send an MP3 for mastering?
You can send an MP3 only if no full-quality file exists, but it is not ideal. MP3 is lossy, so the master cannot recover audio information that was removed during encoding. A WAV or AIFF premaster is the better source.
Should I remove the limiter before mastering?
Remove preview limiters, clippers, and loudness maximizers unless they are an intentional part of the approved mix sound. If the limiter is only there to make the rough bounce loud, bypass it and send a separate loud reference if needed.
How much headroom should a premaster have?
A safe target is peaks around -6 dBFS, but the exact number matters less than avoiding clipping and leaving room for mastering. Do not normalize or lower a clipped file after export; fix the level before bouncing.
Should I send stems or a stereo mix for mastering?
Send the stereo mix for normal stereo mastering. Send stems only if stem mastering was requested or approved. Stems need to line up perfectly, sum to the approved mix, and be clearly labeled.
What should I check before uploading my mastering file?
Play the entire bounce outside the DAW, confirm it is the approved version, check the start and ending, verify sample rate and bit depth, confirm no clipping, bypass preview loudness processing, and use a clear filename.





