Delivery and export
A written export law per partner, so what lands is identical whoever pressed render.
The short version
Is this for you? For exports that vary by editor and files named without a system. Not if one person exports everything to one place; a written export law earns its keep once more than one hand, or more than one destination, is involved.
What leaves your plate. Each partner's export law written down once, the bit depth and render quality kept even at the time they cost, and file names built for whoever goes looking a year from now, not for whoever saved them today.
What starts working. Files that arrive in the shape the platform wants and the team can sort, identical whoever rendered them, with a year of retrievable history behind them, the part of delivery nobody asks for until the day they need it.
The smallest way in. A free audit: three fields, and a written reading back, usually within about three working days. It commits you to nothing.
What to read next. The five levels to see where this fits, or the casework to see it done.
Why the export settings are written down per partner
Every partner's export law is a document
Resolution, frame rate, codec, pass count, target and ceiling bitrates, audio codec and sample rate, and what happens to the metadata: all of it written out. The reason is dull and complete. A preset is something one editor has. A written law is something a studio has, and only the second one survives that editor being on holiday. Each law is BUBI's tested default for that brand as at the date on the document, set against the platform's current recommended upload settings and revised when the platform moves them and when the brand's own needs move, because an export spec written once and never revisited drifts away from what the platform, and the partner, actually want.
The bitrates genuinely differ, and that is the argument
One partner's ceiling is sixty megabits per second. Another's is sixty-five. A third exports the same resolution somewhere between thirty-five and forty-five. Those are decisions about what each platform does to a file and what each brand's footage actually needs, taken against the platform's live guidance and revisited when it changes rather than set once and forgotten. Using the highest number everywhere is the easy answer, and it would cost the third partner render hours for nothing.
Maximum depth and maximum render quality stay on, for different reasons
The two do different jobs, and the copy used to blur them. Maximum depth renders in a higher bit depth, and that is what holds a slow gradient together instead of letting it band, the one artefact viewers notice without knowing the word for it. Maximum render quality is about scaling: it uses a better resampler when a frame is resized, so a downscale or a reframe stays sharp rather than going soft or aliased, and on a GPU-accelerated export that better resampler is often applied for you anyway, so the box mainly earns its place on a software render. What costs time is not the two boxes: the render time belongs to the machine and the project, and a finished long-form export takes around an hour and a half on the current machines and took four hours before we replaced them. It gets paid because banding and soft rescales are both the kind of fault nobody can name and everybody sees.
File names are built for whoever goes looking
Two partners sort by subject and format, so their files read as the name of the piece followed by the version. Another sorts by date and product code, so its files open with the date. Identical craft, opposite convention, and both are correct, because those two teams go hunting for a file in completely different ways.
The same craft, three different sets of numbers
| Resolution, all three | 4K, 3840 × 2160 |
|---|---|
| Frame rate | 25 fps |
| Video codec | H.264, VBR, two pass |
| Lesson channel | target about 45 Mbps, ceiling about 60 |
| Recipe channel, one | target 45 Mbps, ceiling 65 |
| Recipe channel, two | target 35 to 45 Mbps |
| Audio | AAC, 48 kHz, stereo, 384 kbps |
| Master meter | no clipping; limiter ceiling about 1 dB below the top, as insurance |
| Render settings | maximum depth and maximum render quality on, metadata off |
| Vertical formats | 1080 × 1920 at 9:16; feed posts 1080 × 1350 |
| Typical render | about 1.5 hours per long-form export |
What happens after the render
One shared platform, agreed at the start
Whichever one the partner already uses. Everything delivered goes to the same place every time, and that place is reachable by every one of us working on the brand and by nobody outside it. The point of agreeing it in week one is that nobody ever has to ask where something went.
A tracked list, with the states written on it
Every deliverable sits on a shared sheet carrying its state: in progress, waiting on raw footage, delivered, invoiced. It exists so the answer to where is that video is a filter rather than a message, and so a month of work can be looked at as a month and not as a folder.
A correction re-exports one section
When a passage needs fixing, that passage is re-exported instead of the whole file, and the superseded version is renamed where it sits so nobody publishes it by accident. Each video's project and materials stay held for a year after that video is delivered, so a request for an old clip is a search and never a reshoot. The exact windows are set out in the terms.
There is no correct bitrate. There is a correct bitrate for this brand, on this platform, written down where the next person will find it.
Written down, because the setting nobody recorded is the setting somebody guesses at next year.
What changes for you
Files that arrive in the shape the platform wants and the team can sort, identical whoever pressed render, with a year of retrievable history sitting behind them.
Related: Multi-camera editing · Vertical video editing · Colour grading · Mastering for platform
Send the channel and we will read the edits themselves: pacing, cuts, consistency across recent videos. The audit names what post-production would change first, in writing, before anything is committed.
Request your free audit.
Three fields, and a written reading back: a fit snapshot, two or three prioritised opportunities, and a recommended next step.
A person reads the channel and writes the audit by hand: a considered read typically takes three working days. That is the usual shape, not a promised turnaround. We use these details only to reply to you: no lists, no lurking.
What you will get
A fit snapshot: where your channel stands, and whether we are a match.
Two to three opportunities: specific, prioritised, yours to keep.
A recommended next step, even if that step is not us.
The audit is free and commits you to nothing: nobody follows up with a call you did not ask for.
The casework shows all of this done.
The case studies are the same work with the numbers, the sources and the dates attached.
See the casework →