The short version: Platforms in this category genuinely do disappear, and it is worth seeing the evidence rather than treating it as a hypothetical. When we probed 42 platforms this blog has written about, following redirects with a real browser, eight of them were dead or repurposed. One domain now serves an online-casino affiliate site. Another is parked for sale. Three simply fail to resolve. The failure mode is almost never an orderly wind-down with an export window; it is a domain that stops working, often with no announcement at all. Which means the protection has to be in place beforehand, because afterwards there is nothing to log into.
Yes, and here is what we found rather than what we assume.
In July 2026 we probed every platform this blog has named, following redirects with a real user-agent. Eight were gone or repurposed:
That is nearly one in five of the platforms in this space, and none of them announced anything to us. They stopped existing, and their domains were picked up by whoever wanted them, which is what happens to any domain that lapses.
The Castiron case is the instructive one. A vendor who bookmarked their storefront and came back later would land on a gambling site, with a stale banner still advertising the service that used to be there.
Rarely a single dramatic moment. Usually a sequence, and the early parts are visible if you are watching.
Step five is the fork that matters. A company winding down deliberately usually gives notice and an export window. A company that simply stops paying its bills gives you a domain that resolves one Tuesday and does not the next.
You cannot control which you get. You can control whether it matters.
Five things, on the day the domain stops working.
Notice how many of those are conditional on something you either did or did not do beforehand. Your storefront is gone either way. Your customer list is only gone if you never exported it.
That distinction is the entire article: half of what you would lose is optional.
Two things, and they are the more expensive ones.
Search placement. Whatever your storefront ranked for goes with the domain. If you were on a platform subdomain there is no redirect available, so every ranking page becomes a dead link and the placement does not transfer anywhere. Rebuilding it on a new address takes months rather than weeks.
Customer habit. People who learned to order at an address that now shows a casino site do not patiently wait for your announcement. Some will search for you, some will message you, and some will simply buy bread elsewhere.
The second is why the warning signs matter. A vendor who noticed the slowdown and moved deliberately keeps their customers; one who finds out when a customer sends a screenshot does not.
If it happens without warning, in this order.
Step four is the one people forget and it is often the largest recovery. Platforms where you connect your own processor, such as Big Cartel and LocallyGrown.net, mean your payment relationship survives the platform entirely: the money, the history, and the customer payment records are in an account you hold directly.
Step one is the one that saves the business. Customers are forgiving about outages and unforgiving about silence.
Our guide to telling customers about a change covers the tone, and the same principles apply to bad news as to a price rise: say it plainly, say it early, and do not over-explain. Our piece on building a customer email list covers why having somewhere to send that message is the precaution that makes every other one work.
Six, and none require inside knowledge.
None is conclusive alone. Three together over a couple of months is a reason to do your export, check your domain arrangements, and price an alternative, all of which are prudent anyway and cost you an hour.
Worth being fair here: slow support is usually just slow support. Small software companies have quiet periods, and most of them are fine. The point is not to panic at the first delay; it is to have the export already done so that you never need to.
Five things, and together they take about an hour plus five minutes a quarter.
Do those five and a platform disappearing costs you a weekend. Skip them and it costs you the business you spent three years building, which is not an exaggeration when the customer list only existed in one place.
The National Archives' records management guidance is written for institutions and its principle is exactly right here: decide what counts as a record and keep it somewhere you control. The IRS's recordkeeping guidance is the other half, since your retention obligation runs for years and outlasts most platforms' retention policies, let alone their existence.
Somewhat, and less than people assume.
Large general platforms are unlikely to vanish, and the eight we found dead were all small and specialised. So Square, Shopify, Squarespace, and Wix are safe bets on this specific risk.
Small specialised platforms carry more of it, and they are also the ones that do the specialised things: sell-by-weight, drop mechanics, CSA membership management. That is a genuine trade rather than a reason to avoid them.
A solo-operator product carries key-person risk on top. LocallyGrown.net is openly a small operation with support direct from the person who builds it, which is a real strength in responsiveness and a real consideration for continuity.
The honest conclusion is not "choose big." It is that the five habits above make the size question much less important, which is better than choosing a worse-fitting platform out of caution. A vendor with their own domain and a recent export can use a small specialised tool safely.
That matters because the alternative reasoning leads somewhere bad. A vendor who picks a large general platform purely to avoid continuity risk ends up on a tool that cannot express their collection schedule or cap what they can bake, and pays for that every single week for years. The weekly cost of a bad fit is larger and far more certain than the occasional cost of a platform closing.
So weigh continuity as one criterion among ten rather than as a veto, and neutralise it with the habits rather than with the choice.
If you are reading this because your platform has already gone, the sequence is the 48-hour playbook above, plus two things.
Check your payment processor first. Even if the storefront is gone, transaction history and pending funds may be retrievable directly if you held that account.
Do not follow your old domain. As the Castiron case shows, a lapsed domain can be picked up by anyone, and what appears at your old address is no longer connected to the company that was there. Tell your customers the same thing, because the stale branding on those pages is genuinely confusing.
And do the five habits now, on whatever you move to. A vendor who has been through this once and rebuilds without owning their domain or exporting quarterly has learned the wrong lesson from an expensive week. Our guide to tracking income and expenses covers keeping those exports somewhere they will actually be found, which is the difference between having a backup and thinking you do.
If you want to stand something up quickly while you sort the rest out, a trial with your real catalog takes an evening and gets you taking orders again this week.
USA.gov's page on consumer complaints covers escalation routes if money is genuinely owed to you, though for a small subscription the realistic outcome is usually moving on rather than recovering anything.
If you are rebuilding and want somewhere to land quickly, Homegrown is $10 a month billed annually or $12.50 monthly, with 0% commission and 2.9% plus $0.30 processing published up front, and it handles pickup at each place you sell with its own schedule and cutoff, local delivery with a radius and a route, and sales tax calculated, filed, and remitted in all 50 states. The honest bounds: no point-of-sale, no national shipping, no app ecosystem, and no free tier, and it is subject to exactly the same continuity question as anything else here, which is why the five habits matter more than the choice. You can rebuild a catalog in a trial in an evening, which is roughly what recovery should cost if your own folder is in order.
Every platform below shows the same four commercial facts, because a table that lists one platform's transaction fee and not another's is not a comparison. "Not published" means exactly that: the company does not state it publicly.
| Platform | What survives the platform | Subscription (annual) | Free trial | Platform fee | Card processing |
|---|---|---|---|---|---|
| Big Cartel | Your own processor, so payment history is yours | Platinum $12/mo ($144/yr) | 7-day free trial | $0 platform fee | Your own provider, so 2.9% + $0.30 typical |
| LocallyGrown | Your own Stripe account outlives the platform | $0 | n/a, free to start | 3% commission after the first $15,000 | 2.9% + $0.30 processing, your own Stripe |
| Shopify | Large and unlikely to vanish, export anyway | $29/mo Basic | 3-day trial, then $1/mo for 3 | 2% platform fee if not on Shopify Payments | from 2.9% + $0.30 processing |
| Square Online | Large, and the money sits with Square | Free tier; paid from $29/mo per location | 30-day trial on paid plans | $0 platform fee | 3.3% + $0.30 free tier, 2.9% + $0.30 paid |
| Homegrown | Small and specialised, so export quarterly | $10/mo billed annually | 7-day free trial | $0 platform fee (0% commission) | 2.9% + $0.30 processing |
| Cococart | Retention after closure not published | $19/mo store, $59 Pro | 7-day free trial | Platform fee not published | Processing not published |
Yes. Probing 42 platforms this blog has written about in July 2026, eight were dead or repurposed: one domain now serves an online-casino site, one is parked for sale, two redirect to unrelated businesses, and three fail to connect entirely.
Not an orderly wind-down. Support slows, updates stop, communication thins, and then either an announcement with an export window or a domain that simply stops resolving one week.
Your storefront and every link to it, orders in flight, and access to your customer list and catalog if they existed only in their system. About half of that is optional depending on whether you exported beforehand.
Tell customers before you have a solution, open a manual ordering channel, fulfil what you can identify from your email confirmations, check your payment processor directly, and stand something basic up within a week.
With your payment processor. If you connected your own Stripe or similar account, transaction history and pending payouts live there rather than with the platform and survive its disappearance entirely.
On this specific risk, yes. The eight dead platforms were all small and specialised. But small specialised tools do the specialised things, and five cheap habits make the size question much less important than choosing a good fit.
Owning your domain, for about $15 a year. It is the difference between redirecting customers to a new home and having every link you ever shared point at whatever someone else has put there.
This is not hypothetical. Of 42 platforms probed in July 2026, eight were dead or repurposed, including one whose domain now serves an online casino with a stale banner still advertising the old service.
The failure mode is rarely an orderly wind-down. It is a domain that stops working, which means the protection has to exist beforehand, because afterwards there is nothing to log into.
So do the five things, which cost an hour and five minutes a quarter: own your domain, export quarterly with consent status, keep original photographs in your own folder, maintain a direct channel to customers, and know your alternative. Do them and a platform vanishing costs you a weekend. Skip them and it can cost you the customer list you spent three years building.
