Download websites that use Estore Compare

Split testing is one of the quieter signals a storefront can give off. A shop that installs Estore Compare has decided its pages are worth arguing about, which usually means somebody is buying traffic and wants a landing page to earn that spend back instead of trusting an internal opinion about which layout works.

Estore Compare is an A/B testing tool published by a Japanese ecommerce services company and sold as a paid plan with a metered element tied to session volume. Installation is a tag on the page rather than a development project, and the tool then serves competing versions of a whole page, of one block inside a page such as a headline or a hero image, or of two separate addresses. Allocation between the variants can be set by hand or handed to a bandit algorithm that moves traffic toward whichever version is winning while the test is still running, and the vendor pages stress measuring per visitor rather than per session so that a returning shopper keeps seeing the same variant.

Our catalogue files Estore Compare under A/B Testing, which describes the market it was sold into rather than what the script does on the page. Our record also carries Open Source as the license and PHP as the implementation language. We hold no vendor entity, no release history and no version breakdown, even though the product page names a publisher, and stating the gap is better than filling it with a guess.

As of October 2026 we detect Estore Compare on 4,085 sites. Detection depends on the tag appearing in a page we fetch, which sets the natural limits of the count. A test running on a checkout step, an inner product page or a campaign address a crawl never reaches will not register from the home page alone, and a merchant who finished testing and pulled the tag drops out of the next export. The address we hold for the product now forwards to the publisher's main company site, so some share of what we detect is a tag that outlived the campaign that installed it.

The list is worth downloading when your question is about a population rather than a single domain. Testing tools sit where advertising budgets sit, so a set of domains running one is a set of merchants who buy traffic and measure what it does, which is a more specific group than merchants in general. Agencies selling landing page work, competing optimization vendors, and researchers measuring how testing spread through smaller stores all need the whole set at once. Checking one domain answers whether that shop uses it. Only the full export answers who does.