What happens to your work when a platform shuts down

this is a maintenance question, not a doomsday one

The scenario worth planning for is boring rather than dramatic. A platform is acquired and the new owner narrows what it does. A free tier is retired and the archive goes behind a price you were not planning for. A feed changes what it surfaces, so work that is not recent stops being reachable even though it is technically still there. A company simply stops, with a wind-down notice and a date.

None of that requires anyone to behave badly, which is why "pick a platform that will not fail" is not a usable strategy. Every platform is somebody's business, and businesses change shape. The useful question is not which one lasts — it is what you would still hold if any given one stopped tomorrow, and how much work it would be to rebuild the rest.

That question has a cheap answer if you ask it early and an expensive one if you ask it afterwards, which is the entire reason to write it down now.

what actually disappears, and it is not the images

Most artists have the image files. They are on a drive, in a phone backup, in an email to a printer. The loss is almost never the picture — it is everything that made the picture legible to somebody else.

The year the work was made. The exact dimensions, with units. The medium written specifically enough to be searchable. The edition, or the fact that there is not one. Who took the photograph and how they want to be credited. Which show it was in, in which city, in which month, and whether that show was solo or group. The written description of what the image shows. The order the works belong in, which is a piece of information that exists nowhere except in the arrangement itself.

And the links. Every URL you put in a grant application, a wall label, a press release or an email to a curator was a promise that it would still resolve. When the platform goes, they all break at once — including the ones attached to opportunities that have not come back around yet.

The images are the part you can reconstruct. The metadata is the part you cannot, because it was never anywhere but in the platform and in your memory, and memory loses dimensions and dates first.

what portfolio builders are genuinely good at

Website builders — the ones artists actually use — are very good at a real job: presentation you control on a domain you own. You decide the sequence, the pacing, the typography, the amount of white space. For a lot of practices that framing is part of the work rather than packaging around it, and a builder is the only place you fully get it.

The ownership story is also genuinely stronger than a feed's. If the domain is yours, you can move the site and keep the address, which means the links in other people's documents keep working. That is the single most valuable thing a personal domain buys you, and it is worth more than any template.

Their limits are the ordinary ones. You maintain it or it decays, you pay for it indefinitely, and a redesign or a migration can silently renumber every URL — the same link-rot failure, arriving from inside the house. They are also close to invisible to anyone who does not already know your name: a single-artist site has no reason to appear when somebody is looking for a discipline, a material or a city rather than for you.

And they are a different category from an archive, which is the honest thing to say plainly: artlito is not a website builder and does not replace a personal domain. The word between them is "and", not "versus".

what portability actually means

Portability is not "I can download my images". It is whether the structure comes with them. A folder of files named IMG_4471 is not a portable archive; it is the raw material for the reconstruction you were trying to avoid.

The properties worth checking, on anything holding your record:

• Structured fields, not a caption. Title, year, medium, dimensions, edition, rights and exhibition context stored as separate fields — because a caption cannot be exported into anything that understands it.
• An export you have actually run. Not a documented export feature: one you have downloaded and opened, so you know what is in it before you need it.
• Stable addresses. One canonical URL per work that is not expected to change, so the links you hand out keep their promise.
• A visual description written for each work, which is both what a screen reader announces and the only reason an image is findable in words at all.
• Rights and credit recorded alongside the work rather than remembered — including the photographer credit, which is the one that gets lost.
• Your own copy of the full-resolution files, in a place that is not any platform. This one is unglamorous and it is the floor everything else stands on.

what artlito does about this, and what it does not

artlito is built as a structured record rather than a feed. Each work is catalogued in the fields a work actually has, and each one has a permanent URL that is meant to still resolve when somebody opens it from a grant application eighteen months later. Permalink stability is treated as a requirement rather than a nicety, and the browse is structural — by discipline, by medium, by role — so work can be found by what it is rather than by whose name is already known.

There are no follower counts, no likes, no trending page and no engagement-ranked default ordering. That is a design decision rather than a missing feature, and it is relevant here: a platform organised around recency has a structural reason to bury an archive, and one organised around the record does not.

The limits, stated plainly. artlito is not a website builder, as above. There is no marketplace payment flow — you cannot sell a work through artlito today; it records what exists and who made it, and a sale happens wherever it already happens for you. The audience is smaller than any large feed's and always will be, because a structured archive is not a reach machine.

And the honest one, which an article on this subject cannot skip: artlito is itself a platform, run by a company, and the argument above applies to it too. It is not exempt from its own reasoning. What that obliges us to do is make the record exportable and the addresses stable, so that the answer to "what if artlito stops" is a file you already hold rather than a promise about our intentions. Any platform telling you it is the one that will definitely still be here is asking you to trust a thing nobody can actually know.

three layers, and what each one owes you

The arrangement most working artists arrive at is not one tool. It is three, each doing a job the others do badly — and the failure mode is asking one of them to do all three.

if you only do one thing

Write down the metadata for the work you made this year, while you still have it to hand. Not the whole back catalogue — this year. Title, year, medium written specifically, dimensions with units, edition or a statement that there is not one, where it was shown, and who photographed it.

That single act is what makes every later move cheap. It survives any platform decision, it is what an export needs in order to be worth exporting, and it is the part that becomes genuinely hard to reconstruct after about eighteen months. Everything else in this piece is a variation on having done it.

The failure worth naming is the common one: a decade of posting, and nothing at the end of it that a registrar could check. Ten years of posts is not an archive. It is ten years of posts, and the dates went first.