Music
Mastering is not making it louder, and the confusion costs musicians money
The final stage before release is small, technical and widely misunderstood. It is the difference between a mix that works everywhere and one that only works where it was made.

Comparisons of the mastering stage usually pick a winner. This one picks the circumstances, which is more useful.
The difference in one place
- Mastering treats a finished stereo mix, so its adjustments are broad rather than surgical.
- A large part of the job is consistency across a release and correct delivery formats.
- A mastering engineer is valuable partly for being the first outside ear on the material.
What the stage actually involves
Mastering starts from a finished stereo mix, meaning the individual instruments are no longer separable and cannot be adjusted independently. The available tools are broad: equalisation across the whole mix, compression across the whole mix, level, and stereo width. The engineer is aiming for a version that holds together on a phone speaker, in a car and on studio monitors, none of which agree.
That is a translation problem rather than a creative one, and it is why mastering rooms are built and measured so carefully. The changes are usually small, which is exactly why people underestimate how much difference they make.
Consistency across a release
On an album, mastering makes separate mixes recorded in different places over months feel like one body of work. That includes level matching, tonal matching and setting the gaps between tracks so the sequence flows at the intended pace.
Gap timing is a genuinely musical decision, and a badly judged pause can undo a transition the artist spent weeks constructing. Crossfades and segues are also handled here, which is why the mastering engineer effectively determines how an album is experienced. Single-track releases lose most of this work, which is one reason the stage is now more often treated as a formality.
Delivery is half the job
Different destinations need different files, and a master delivered for streaming is not identical to one prepared for vinyl cutting. Vinyl imposes real physical constraints, including limits on bass energy in the stereo sides and on sibilance, which can require a distinct master. Streaming platforms apply loudness normalisation, so a master must be prepared with that processing in mind rather than against it.
Metadata, track identifiers and gap information all get embedded at this stage and are tedious to correct afterwards. Getting delivery wrong produces problems that appear only after release, which is the most expensive time to discover them.
The value of an outside ear
By the time a mix is finished, everyone who made it has heard it hundreds of times and has stopped hearing it accurately. A mastering engineer hears it once, in a neutral room, without knowing what it was supposed to sound like. That single fresh judgement catches problems that months of familiarity have rendered invisible to everyone else.
Structurally, it is a substantial part of what is being paid for, and it is the part that automated services cannot supply.
Many experienced producers say they book mastering primarily for the opinion and secondarily for the processing.
Automated mastering and what it can do
Algorithmic services analyse a mix and apply corrective processing based on comparisons with reference material. For straightforward material aiming at a conventional sound, the results are frequently adequate and cost almost nothing. What they cannot do is recognise that an unusual balance was intentional, so deliberately odd mixes get normalised toward the average.
In the mix, they also cannot make sequencing judgements or advise that a mix problem should be fixed at the mix rather than patched afterwards. Used knowingly they are a reasonable tool; used as a substitute for judgement they flatten exactly what made a record distinctive.
Budget and streaming figures are rarely audited and frequently strategic.
What mastering cannot repair
A mix with a buried vocal cannot be fixed at mastering, because the vocal is no longer a separate element to raise. Phase problems, poor arrangement and unbalanced low end all have to go back to the mix, and a good engineer will say so. Anyone promising to fix such issues at the final stage is either misdescribing the work or planning to disguise it with heavy processing.
Structurally, the correct response to a mastering engineer requesting a revised mix is relief rather than irritation. Understanding the boundary saves money, since it stops people paying for a stage that cannot solve their actual problem.
Side by side
| Consideration | What it means in practice |
|---|---|
| What the stage actually involves | Mastering treats a finished stereo mix, so its adjustments are broad rather than surgical. |
| Consistency across a release | A large part of the job is consistency across a release and correct delivery formats. |
| Delivery is half the job | A mastering engineer is valuable partly for being the first outside ear on the material. |
The takeaway
Mastering makes a mix work everywhere; it cannot make a mix work at all.
The premise gets you in. The structure decides whether you stay.
Questions readers ask
Do I need mastering for a single track?
For a release, yes, mainly for translation and correct delivery. The album-level consistency work simply does not apply.
Can the mixing engineer master their own work?
Technically yes, and many do. They lose the fresh-ears benefit, which is arguably the largest single advantage of the separate stage.
Also by Sanjeev Pillai
- Records got louder for thirty years, and the reason was competition rather than tasteMusic
- A record producer is doing four different jobs and the credit does not distinguish themMusic
- The vocal you love was assembled from thirty takes, and that is normal practiceMusic
- Album sequencing still matters even though almost nobody listens in orderMusic





