How PageFleets works
PageFleets shows you how a website looks and loads from several parts of the world at once. The same page can render differently depending on where the visitor is — a nearer or farther CDN edge, a geo-targeted redirect, a different default language, sometimes different content entirely. Checking from your own machine only ever tells you about your own location.
What happens when you preview a URL
- You enter a URL. Any public website, at any viewport from mobile to wide desktop.
- We render it in each region. A real headless Chrome browser running in that region loads the page and waits for it to settle — not a simulation, the actual page as a local visitor would receive it.
- We capture the full page. The whole scrollable page, not just the first screen, plus a latency breakdown — DNS, TCP, TLS and time to first byte — measured from the region itself.
- You compare regions side by side. Line the captures up against each other to spot what changes from one place to the next.
What it is good for
- Confirming a CDN edge is actually serving in a region, not falling back to a distant origin.
- Catching geo-targeted redirects, consent banners and "choose your region" interstitials before your users do.
- Verifying localized content and language detection render the way you intended.
- Seeing real first-paint latency from where your audience is, instead of guessing from your own connection.
- Checking a responsive layout at a specific viewport, from mobile to wide desktop.
Where captures run from
Every capture runs from a real node in the region you pick:
For developers
The same capture is available over a JSON API, so you can wire multi-region screenshots into your own checks and pipelines. See the API documentation.