Homepage Snapshot
Paste a web address and get a full-page screenshot of the homepage — fonts loaded, lazy images in place, cookie wall cleared.
How it works
The page is opened for real
A headless Chrome loads the homepage with a genuine user agent and viewport — no proxy render, no cached thumbnail.
It is given time to finish
Network activity settles, web fonts resolve, the whole page is scrolled so lazy images load, and the consent wall is dismissed — all inside one deadline, so nothing runs away.
Then it is captured
Full page or viewport, at 1x or retina 2x, saved as PNG, JPEG or WebP with the page title and final URL alongside it.
What a good snapshot needs
Anyone can point a headless browser at a URL. The difference between a usable screenshot and a grey skeleton is what happens in the two seconds before the shutter.
- Lazy images forced to load instead of rendering as empty boxes
- Web fonts resolved, so the type is the real typeface
- Cookie and consent walls clicked away, then any leftover overlay removed
- Sticky headers captured once, not repeated down the page
- Device emulation with touch and a mobile user agent, not a narrow desktop
- Runaway infinite-scroll pages capped instead of failing
- Bot walls retried, and a slow page captured as it stands rather than failing outright
Good for
Design archives and competitor sweeps, portfolio thumbnails, before/after records of a redesign, client proposals, bug reports, and link previews that need to look like the real site.
Captures are generated on demand. An identical repeat is answered from a 10-minute in-memory cache; nothing is written to disk and nothing is kept after that.
Questions
How do I take a screenshot of a website homepage?+
Paste the address, pick a device, and press Take snapshot. A headless Chrome opens the page on our server, waits for the network and web fonts to settle, scrolls the page so lazy images load, clears the cookie banner and then captures it. You get the image back in a few seconds and can download it as PNG, JPEG or WebP.
Why do other screenshot tools return half-loaded pages?+
Because they capture as soon as the page fires its load event. Modern sites load their images only when you scroll to them and swap their web fonts in later, so a shot taken at load time shows grey boxes and the wrong typeface. This tool scrolls the entire page first, flips lazy images to eager, waits for document.fonts.ready and only then takes the picture.
Does it remove cookie banners?+
It tries, and usually succeeds. The known consent handlers (OneTrust, Cookiebot, Didomi, Usercentrics, Quantcast and others) are clicked by their own button, then any button reading "Accept all", "I agree" or similar is used as a fallback, and whatever full-screen layer is left over is removed. Switch Clear pop-ups off if you want the page exactly as a first-time visitor sees it.
Can I capture the mobile version of a site?+
Yes. The tablet and mobile presets emulate a real device — touch support, a mobile user agent and a 2x pixel ratio — so responsive sites serve their mobile layout rather than a squeezed desktop one.
Can it capture a page other than the homepage?+
Yes. By default the address is reduced to the site root, since that is what most people want. Turn on Exact URL to capture the precise path you pasted, including query parameters.
How tall can a full-page capture be?+
Up to 20,000 pixels. Past that the result is capped and labelled, because browsers cannot reliably rasterise an infinite-scroll page and the resulting file would be unusable anyway.
Why do some sites return an error instead of an image?+
A handful of publishers answer datacenter traffic with a bot wall rather than their homepage. The capture is retried once, and if the site still answers with an error code but has rendered a real page underneath it, the image is returned anyway with a note explaining what happened. Only a genuinely empty response, a missing domain or a dead server produces an error now.
Why is it faster than it used to be?+
Three reasons. The browser is kept warm between captures instead of being launched from scratch each time. Analytics and session-replay beacons — which paint nothing but keep the network busy — are dropped by default. And every wait is bounded by one shared deadline, so a page that never stops loading is photographed as it stands rather than running until it times out. Most homepages now finish in two to six seconds.
Can I remove the ads from the screenshot?+
Yes, switch on Block ads. Ad exchanges and content-recommendation widgets are then refused, which makes the capture faster and the page cleaner. It is off by default because the result is no longer exactly what a visitor sees — invisible analytics beacons are always blocked, but anything that paints is left alone unless you ask.
Is there an API for website screenshots?+
Yes. GET https://convertfleet.com/api/homepage-snapshot?url=stripe.com returns JSON with the image plus the page title and final URL, and adding &response=binary returns the raw image bytes. Every option on this page is available as a parameter.
Build with our Homepage Snapshot API
Integrate our high-speed conversion engine directly into your own applications. Get free api keys for website screenshot api and more.