Download websites that use Shopify

About Shopify

Detecting a hosted platform means something different from detecting installed software, and the difference changes how every number on this page should be read. As of September 2026 we see it on 3,503,354 domains, and each of those represents a decision to not operate infrastructure rather than a decision about which infrastructure to operate.

That framing matters because it tells you what kind of company is on the other end. Merchants on a hosted platform have outsourced the parts of the problem that generate the most technical decisions, which means the remaining decisions cluster in a narrow and predictable band: apps, themes, payments, shipping, marketing tooling. If you sell into that band, the population is unusually well qualified. If you sell infrastructure, it is almost entirely unqualified, and knowing that in advance saves a great deal of wasted effort.

The country column needs an explicit warning. Canada accounts for 3,240,088 of detections, which looks like a statement about where the merchants are and is not. On a hosted platform the location we observe follows the platform's own serving infrastructure rather than the trading address of the business. Anyone building territory assignments from that column will allocate a global merchant base to a single country and wonder why the outreach fails.

The .com zone carries the largest share, and that one is closer to informative. A merchant base that buys a commercial domain before it buys anything else produces exactly this distribution, and the exceptions are worth looking at.

Because the platform is uniform, the discriminating signals sit around it rather than inside it. Which stores carry a particular analytics account, which sit behind a given proxy, which run a recognisable theme family, which appeared since the previous rebuild. That last one is the closest thing this space has to an intent signal: a store that did not exist last week is a store that is buying things this week.

None of those are answerable from a directory, because they are all intersections and diffs rather than lookups. They need a full population you can hold, filter and compare against its previous state.

The webatla export provides that. Every domain we detect on the platform, with the rest of the identified stack on the same row, the hosting country and the TLD, delivered as a single file and rebuilt daily so that comparing two days is a local operation.

The store lifecycle is worth understanding before building anything on this data. Hosted commerce has a high creation rate and a high early failure rate, so any snapshot contains a substantial tail of stores that were set up and effectively abandoned. Filtering on the presence of an analytics account, a custom domain or a payment integration separates the operating stores from the parked ones, and doing that first will change the economics of any outreach built on top.

Theme and app detection is where the remaining discrimination sits. Because the platform itself is uniform, what a merchant has added to it is close to the only technical evidence of their maturity and priorities. A store carrying a subscription app, a reviews widget and a translation layer is at a different stage from one running an unmodified default theme, and that difference is visible in the same row of the export rather than requiring a separate lookup per store.

Shopify Inc. runs the platform commercially, which is what the Proprietary label reflects, and it has been operating since 2006.