Uploading drawn animation to a general video platform tends to disappoint everybody involved, and the reasons are structural rather than a matter of quality. These platforms were built around a set of assumptions about what a video is, and animated illustration violates most of them at once.

It is worth setting out plainly, because a great deal of frustration comes from artists assuming the poor result reflects on the work when it usually reflects the fit between the work and the machine it was fed into.

Duration-weighted metrics punish short work

Most recommendation systems reward total watch time, which favours long content. A ninety-second animated piece can be watched to completion by everybody who starts it and still be treated as less valuable than a twenty-minute video that most people abandon halfway. The metric is not measuring quality; it is measuring minutes.

This is close to unfixable from the creator side, since drawn animation is expensive per minute, as covered in the piece on what the field covers. Producing long-form work to satisfy a metric usually means producing worse work slowly.

Classification handles drawing inconsistently

Automated content classification is trained overwhelmingly on photographic material, and its behaviour on drawings is erratic. Work gets flagged that should not be, and occasionally passes that should not, and the inconsistency is worse for stylised material because it sits furthest from the training data.

The practical result is unpredictable enforcement: two similar uploads receive different treatment, appeals go to systems that cannot explain themselves, and nothing about the outcome teaches the creator what to do differently next time.

Compression and presentation

General platforms encode aggressively for bandwidth, and drawn work suffers characteristically, with banding in flat areas and haloing on lines, as described in the piece on delivering animation. Work that looked clean on export arrives softened and stepped.

Presentation compounds it. Autoplay, thumbnails chosen by algorithm, and surrounding recommendations all frame the work in a context nobody chose, and for illustration where framing carries meaning, that loss of context is more damaging than it would be for other material.

Credit and discovery work poorly

Attribution on general platforms is a channel name and a description that most viewers never read, and re-uploads are common and hard to police. Discovery is driven by recommendation rather than search, so an artist has little influence over who arrives and no relationship with them afterwards.

This is the same attribution economy described in the piece on why work appears everywhere at once, with the additional problem that the platform itself is the intermediary rather than a scraper.

The context problem

There is a subtler cost that rarely gets named. A general platform places work in a stream alongside whatever its system thinks is similar, and for drawn material that neighbourhood is frequently a poor match in tone, quality and intent. Readers form an impression of a piece partly from what surrounds it, and an artist who has thought carefully about how their work addresses a viewer loses that control entirely at the moment of publication.

That loss is not measurable and it is real. Work that reads as considered on a personal site can read as disposable in a grid of thumbnails, and nothing about the file has changed. Choosing where to publish is therefore part of the editorial decision rather than a separate distribution chore that happens after the work is finished.

What works better

Specialist platforms, personal sites, and direct distribution all handle drawn work better, because they were built for audiences who came deliberately rather than for a recommendation engine. They are smaller, and smaller audiences that arrive on purpose convert far better than large ones that arrive by accident.

The general rule follows from the resilience argument in the piece on why publishability varies: use platforms for distribution, keep the canonical version and the audience relationship somewhere you control.

Choosing where animated illustration belongs

None of this means never uploading anywhere general. It means expecting the fit to be poor, not measuring the work by results from a system built for something else, and putting the effort into channels where short, made-on-purpose drawn work is what the audience came for.

Characters here remain fictional and unmistakably adult wherever the work is published, as set out in the reader agreement.