Data-driven research, statistics, and analysis on the technologies, infrastructure, domains, and trends shaping the web.
11.3% of websites carry Cloudflare's analytics beacon in their HTML, more than carry a Google Analytics 4 tag there.
What websites load and where it comes from, which fonts, libraries, markup and analytics their homepages carry, and the web design companies they link to.
Which certificate authorities secure the web, how much of it is free, who really picks the certificate, and how many websites still serve an expired one.
Which languages, licenses, companies and developers lead the 480,428 GitHub repositories with 100 or more stars, and how quickly AI is taking over the new ones.
Which providers the web actually resolves to, how many domain names sit on a single IP address, who runs the DNS, and how much of it points at nothing.
How many domains exist on each extension, how many are real websites, how many are held as inventory, and what the working ones are actually built with.
Which languages websites claim, how many claim none, where in the page the answer sits, and why a .com and a .de behave like two different internets.
How many websites run a CMS, which platforms build them, what the rest of the web is built with, and why the platform that builds the most websites is not the platform that builds the most important ones.
How many domains publish SPF and DMARC, what those records actually say, where the reports go, and which mail providers authenticate their customers. Counted across every domain on the internet with an MX record.
How many websites have a robots.txt, what they allow and block, which crawlers are blocked and named most often, and which platforms wrote the rules. Counted across the entire internet, not sampled.
How many websites block GPTBot, ClaudeBot, CCBot and the other AI crawlers, which of the biggest sites do it, and why most of the blocks trace back to one vendor's default. Counted across the entire internet, with named sites.
How many websites have an llms.txt, which platforms generated most of them, what they contain, where their links point, and why the most quoted adoption figure counts placeholders and templates rather than decisions.
Figures may be reproduced with attribution to StackScan. Atom feed