Slow Page Example: A Deliberately Slow Response
My greyhound whelped in the middle of a hare chase, and the litter joined the hunt and caught it; this page has yet to arrive.
The short answer
- What it is
- A slow page is one that takes too long to arrive and finish loading. Here, a page counts as slow when fetch time plus render time is over 3,000 milliseconds.
- Why it matters
- People leave pages that make them wait. Slow responses also lower how much a search engine will crawl, and loading speed feeds into Core Web Vitals.
- How a crawler should report it
- Report the URL as a slow page when fetch time plus render time is over 3,000 ms, and show both numbers in milliseconds. Do not raise it for a page that takes 2.5 seconds in total.
- Live examples
/lab/slow-pages/(positive)/lab/slow-pages/near-miss/(near miss)
How a long wait for the page affects people and bots
A slow website example is easy to build: make the server wait before it sends the page. Real sites get slow for less simple reasons. Database queries may lack an index. Pages may be built fresh for each visit with no cache. A shared host may be short on memory. A page can also arrive quickly and then spend seconds running large scripts and third-party tags before it is ready to use.
Browsers show a blank tab until the first bytes arrive, and then they paint what they can. Google's Core Web Vitals call a Largest Contentful Paint of 2.5 seconds or less good. They also use field data from real Chrome users. On top of that, Google adjusts its crawl rate based on how quickly and reliably a server replies. A slow server gets fewer requests, so new pages can take longer to be found.
Measure the two parts apart from each other. The time to first byte shows the server side, and curl -w "%{time_starttransfer}" gives you that number. The network panel in browser tools shows what happens next, with each file on a waterfall. A crawler should record the fetch time and the render time for each URL. Run tests more than once, since a single reading can be off because of a cold cache.
To fix server delays, add page caching, tune the slowest queries, and use a CDN for static files. To fix browser delays, cut script size, delay tags that are not needed right away, and compress images. Load the main image early rather than late. After each change, test the same URLs again. Compare your results with field data, since lab tests cannot show what every real visitor gets.