Once your mix is finished, sending it to a mastering engineer should be straightforward. In practice, unclear file names, missing versions and conflicting mix exports can create unnecessary delays before mastering even begins.
The mastering engineer should not have to guess which file is final, whether the limiter was intentional, where the fade should end or which version belongs on the release.
A well-prepared delivery package gives the engineer the information needed to begin confidently while keeping the focus where it belongs: making the music sound its best.
This guide is not about how to mix your song before mastering. It is about exactly what to send, how to organise it and why each part matters.
For traditional stereo mastering, the most important file is your approved final mix.
This should be the version that represents the balance, production and emotional direction you want to preserve. Mastering can improve translation, tonal balance, dynamics and overall presentation, but it should not be used to decide between unfinished mixes.
Your final mix should generally be supplied as:
There is no need to convert a 48 kHz session to 44.1 kHz before sending it. The mastering engineer can create the required release formats from the original high-resolution file.
The same applies to bit depth. Send the highest-quality version available rather than reducing it prematurely.
Normalisation is not necessary before mastering. It simply changes the file’s overall level and provides no benefit to the engineer.
Dither is also usually unnecessary at this stage. It is most useful when reducing a file to a lower bit depth, which generally happens during final mastering and delivery.
Sending a 24-bit or 32-bit float mix without added dither gives the mastering engineer the cleanest starting point.
There is also no universal rule saying your mix must peak at exactly -6 dBFS. The important thing is that it is not clipping unintentionally and has enough room for the dynamics of the song to remain intact.
A healthy, unclipped mix is more useful than one that has been turned down purely to reach an arbitrary number.
This depends on what the limiter is doing.
If it was added only to make the mix louder for comparison or presentation, it is usually best to send a version without it. Heavy limiting can restrict the mastering engineer’s ability to shape transients, dynamics and overall movement.
However, not every mix-bus processor should automatically be removed.
Bus compression, saturation, EQ or limiting may be an important part of the mix’s character. If removing it changes the balance, energy or emotional impact of the track, it may be better to leave it in place.
In that situation, send both:
Name the files clearly and tell the mastering engineer which version represents your artistic preference. This gives them context and flexibility without forcing them to reconstruct the sound you already approved.
A mastering engineer should be able to identify the correct file without opening several nearly identical versions.
Avoid file names such as:
Instead, include useful information in the name:
Version numbers or dates can also help when several updates have been exchanged:
Artist – Song Title – Mix 03 – 2026-08-04.wav
Clear file naming reduces the chance of the wrong version being mastered and creates a reliable record of what was supplied.
A reference mix allows the mastering engineer to understand how you have been hearing the song.
This might be:
The reference is not necessarily the file that should be mastered. It is there to communicate intent.
For example, your clean mix may have more headroom, while the reference version demonstrates the density, aggression or brightness the artist has become accustomed to.
Providing both helps the engineer avoid taking the master in a technically impressive but artistically inappropriate direction.
If the release requires alternate versions, it is best to send them at the same time as the main mix.
These may include:
Mastering these versions together helps maintain consistency across the entire release.
Do not assume an instrumental can always be created after mastering by simply muting the vocal. Removing a lead vocal can change how the mix bus reacts, particularly when compression, limiting or saturation is involved.
Each alternate should be exported properly from the mix session and clearly identified.
Stem mastering is different from sending a collection of individual multitracks for mixing.
For stem mastering, the song is usually divided into a small number of logical groups, such as:
All stems should begin at the same point, finish at the same point and remain perfectly aligned.
When combined at unity gain, they should reproduce the approved stereo mix as closely as possible. If the stems do not sum back to the mix, something may be missing or the routing and processing may have changed during export.
You should also include the approved stereo mix so the mastering engineer has a clear reference for how the stems are intended to work together.
Stem mastering can provide additional control, but it should not be treated as an unstructured mix revision. The files still need to represent a finished production.
A short set of notes can help the mastering engineer understand the project.
Useful information includes:
Try to describe the musical result rather than prescribing exact technical moves.
For example:
“I want the chorus to feel wider and more open than the verse.”
is generally more useful than:
“Add 1.5 dB at 12 kHz and use a stereo widener.”
Mastering engineers need to understand what you want the listener to feel. The exact processing used to achieve that should remain part of their judgement.
For a single, provide the correct:
For an EP or album, also include:
Not every release format embeds metadata in the same way, but supplying the information clearly prevents spelling mistakes and inconsistent titles later.
It is much easier to confirm these details before the final masters are delivered than to correct them across multiple formats afterward.
When mastering several songs as one project, do not send them as unrelated singles without explaining how they belong together.
The mastering engineer needs to consider:
Include the complete sequence and identify any tracks that should flow directly into one another.
A song that works perfectly in isolation may need a slightly different treatment when placed between two other tracks. Album mastering is about shaping the experience of the entire release, not merely processing each song independently.
Unless your mastering engineer specifically requests them, avoid sending:
More files do not automatically provide more flexibility. A smaller, clearly organised package is often more useful than a folder full of uncertain options.
What you send a mastering engineer affects more than administration.
It gives them a clear understanding of the project, protects the decisions already made during mixing and reduces the chance of mastering the wrong version.
A strong delivery package should contain:
The aim is not to overwhelm the engineer with options. It is to remove uncertainty.
At Moreish Studios, I want the mastering process to begin with a clear understanding of what the artist and mixing engineer have already created. Good organisation makes that possible and allows the work to focus immediately on translation, consistency and the final emotional impact of the music.
If you are preparing a release for mastering, learn more about our Mastering services or get in touch to discuss the best files and formats for your project.
