Download websites that use Font Awesome

About Font Awesome

Icon sets are not usually considered strategic technology, and that is exactly why Font Awesome is useful in this dataset. Nobody agonises over the choice, nobody upgrades it for its own sake, and nobody removes it once a theme depends on it. It is inert, and inert things make excellent timestamps.

As of September 2026 we detect it on 11,951,844 sites. The interesting part is not the total but the release distribution underneath it, because the project went through a well known break in its class naming and delivery model. A site on an older release is a site whose front end predates that break and was never migrated across it, and 4.7.0 is the release we see most often.

That single fact does a lot of work. It dates the build of the site with more precision than a copyright notice, and unlike design fashion it is machine readable at scale. Cross-referenced with the platform above and the interpreter below, it produces a defensible estimate of when a property was last meaningfully rebuilt rather than merely edited.

For an agency, that estimate is the qualification step. A property that is commercially active, running current analytics, and carrying a front-end stack from several cycles ago is a redesign conversation. The same property with a current stack is not. Distinguishing them at scale is the difference between a prospect list and a directory dump.

There is a caveat that keeps this honest. The library is frequently bundled, renamed or served from a theme's own asset path, so the release we observe is less reliable than for software that announces itself in a header. We publish the string we saw rather than normalising it into something tidier, because a clean number that quietly guesses is worse than a messy one that does not.

United States carries 4,417,320 of detections and the .com zone leads, consistent with where template-built commercial sites accumulated fastest.

The webatla export puts the observed release next to every other detection on the same domain, in one file rebuilt daily. Age inference is an intersection problem, and intersections are cheap only when you hold the whole population locally.

The dating inference deserves one more layer of care. Sites are edited far more often than they are rebuilt, so a stale front-end library indicates the age of the build rather than the age of the content. A property with a decade-old asset stack and a blog post from last week is being maintained, just not reconstructed, and confusing those two states will send a redesign pitch to someone who is perfectly happy.

That distinction is recoverable from the data, which is the argument for holding the full stack together. Recent analytics generations, current plugin releases and active security tooling all indicate ongoing attention, while the presentation layer indicates when the shape of the site was last decided. Reading both at once separates neglected properties from merely old ones, and only one of those is a prospect.

Fonticons Inc. publishes it on a Freemium model and it has been available since 2012, distributed widely enough through themes and templates that the vendor's own customer list would capture only a fraction of the real footprint.