Download websites that use Swell

About Swell

Counting a commerce platform honestly gets hard when one product can be run two different ways, and Swell is a clear example of why. A platform that renders its own storefront leaves a signature in the markup on every page. A headless build leaves nothing on the vendor's side. The shop is a custom front end put together by a developer, and the platform is reachable only through an API that the browser may or may not call in a way a crawler can observe.

So the number below should be read as a lower bound. As of October 2026 we detect Swell on 93 sites. The detected set is made up of stores whose front end happens to expose the connection, through a client side call to the vendor's API host, a checkout redirect, or a published storefront token. Stores that proxy every request through their own backend look, from the outside, like ordinary custom sites, and no fingerprint survives for us to match.

Presence carries a weaker implication about the team than it first appears to. The vendor sells the platform on monthly subscription plans that run from a small storefront assembled with prebuilt themes and a visual builder up to tiers priced for large catalogues, so a shop running Swell can be a template build as easily as a bespoke one. Our record carries JavaScript and Proprietary against the entry, and neither field decides which of the two a given domain is. What the fingerprint marks reliably is a merchant paying for a hosted commerce backend, and knowing that much is already useful.

Age is worth a sentence of its own. A platform that appeared in 2019 has no legacy install base, so every domain in the list represents a build or a replatform made in recent years rather than a decision inherited from a decade ago. For anyone studying migration flows, a young platform's install list is closer to a record of recent movement than to a snapshot of history.

The commercial model matters for a second reason. A store detected here runs on the vendor's own infrastructure and pays a subscription to do it, so the vendor's customer list and the population visible on the web are not two independent sets. Our record shows Proprietary against the entry, and whatever that field settles on, the list we can publish is the crawlable part of a roster the vendor holds in full and does not share.

For a researcher the useful move is comparative. Pull the detected domains for Swell and for the platforms it competes with, then compare the two populations on the things a domain reveals anyway: country code, payment providers, front end framework, whether a headless build actually shipped or the brand went back to a hosted cart a year later. Population questions like those cannot be answered one domain at a time, which for a category this small is the difference between an afternoon and a project.