Loading Time

Loading time is how long a page takes to become visible and usable after a visitor requests it. It runs from the request to the point where the main content is on screen and responds to input, and it is measured in seconds.

How search engines treat it

Google measures the visitor’s experience of a page through its Core Web Vitals and states in its page experience documentation that those metrics are used by its ranking systems. The same document says search seeks to show the most relevant content even when the page experience is poor, and that a good experience contributes where many pages are equally helpful. Loading time is a tiebreaker, then, and a real one, but it does not lift a page that fails to answer the query. The individual metrics are described under Core Web Vitals.

Loading time for a crawler

A crawler has no patience and often no JavaScript. It requests the HTML, waits a bounded time, and reads what arrived. For that reader, loading time is the time to the document, and the document has to contain the content. This site’s own crawler, fetching 9,483 small-business sites with scripts off, found 7.2% behind bot protection that served a challenge page instead of the site, and 3.0% whose content rendered only in JavaScript, so the crawler received an empty page. Neither site loaded slowly in a browser. Both served nothing to the machine. The figures are in what 9,483 small business websites serve to an AI crawler.

A common question

Can a page load fast in a browser and still be slow for a crawler?

Worse than slow: it can arrive empty. A crawler asks for the HTML, waits a bounded time and reads whatever came back, often with no JavaScript, so what counts is the time to a document that already holds the content. Fetching 9,483 small-business sites with scripts off, this site's crawler met bot protection serving a challenge page on 7.2% of them, and on a further 3.0% the content existed only in JavaScript, so nothing arrived. None of those pages was slow in a browser.

Sources

Ranking treatment: Google’s page experience documentation. The 9,483-site figures are this site’s own, computed from its audit records; crawler ran without JavaScript, sample drawn per metro and trade.