Download websites that use Kameleoon

An experimentation platform is a statement about how a company makes decisions. Installing one means somebody argued that page changes should be measured instead of debated, and then found the budget to keep measuring. Sites that run testing tools tend to have enough traffic to split meaningfully and a person whose performance review depends on the outcome.

Kameleoon is one of the tools that argument ends in. The platform covers web experimentation, feature flagging and rule driven personalisation, and it reaches the browser as a JavaScript tag that has to load early enough to alter content before a visitor sees the original version. Kameleoon sells it as a Proprietary product, and our record for it begins in 2012. As of October 2026 we detect Kameleoon on 244 sites.

Two forces blunt that number. Experimentation vendors increasingly offer server side and edge execution, where a test is resolved before the HTML leaves the origin and nothing identifiable ever arrives in the browser. Tags delivered through a tag manager rather than a hardcoded script can also escape a first pass crawl. Our count therefore leans towards the classic client side installation, which remains common but is no longer the only shape the product takes. The undercount is not evenly spread either, since the organisations most likely to run server side are the larger ones with engineering capacity.

For a researcher the useful property of this category is that membership is voluntary and expensive. Nobody ends up with an enterprise testing platform by accident, so a domain list of Kameleoon installations sits close to a list of organisations that fund a conversion or growth function. Competing vendors read it as a renewal pipeline. Agencies read it as qualification, because a prospect already running experiments does not need to be sold the idea, only a better execution of it. Analysts read it as adoption evidence in a market where public customer numbers are scarce and vendor claims are unverifiable.

There is a second reading that has nothing to do with sales. A personalisation tag decides what a visitor sees, which makes it interesting to anyone studying how content varies between people on the same URL, how consent frameworks interact with tools that must run before render, or how much of a page's visible text is actually fixed. Work of that sort needs a population to sample from, not a hunch about one site.

Checking a single domain tells you what one company chose. The downloadable list tells you which companies chose it, under which top level domains, in which countries, and how the population shifts between rebuilds. Assembling the same thing by hand means crawling the pages we already crawl and repeating it every time you want a fresh answer. The crawl is the work being bought. The answer to one question was never the expensive part.