What Is a Good Page Load Time? A Clear Target and How to Check It

A good page load time is one where the main content of the page is visible within 2.5 seconds for at least three in four real visits. That is a specific target: it uses Largest Contentful Paint rather than "fully loaded", and the 75th percentile rather than an average. This guide explains why, lists targets for the related timings and shows how to check your own pages.

The short answer

Aim for Largest Contentful Paint of 2.5 seconds or less at the 75th percentile of real visits. In plain words: for at least three in four visits, the main content should be visible within 2.5 seconds.

Why not "fully loaded" time?

"Load time" in many tools means when every resource on the page has finished: images below the fold, chat widgets, analytics scripts. Visitors do not wait for those. A page can feel ready in 1.5 seconds and still report a 7-second full load.

LCP measures when the largest piece of content in view, usually the main image or heading, has appeared. It matches what people experience much more closely.

Why the 75th percentile?

An average mixes fast and slow visits into one number that describes nobody. The 75th percentile says: three in four visits were this fast or faster. It keeps the slow quarter in view without letting a few extreme visits dominate.

Targets for each timing

The table gives the thresholds for the four timings that matter most. LCP is the headline target; FCP and TTFB explain why LCP is slow; INP covers how quickly the page responds once loaded.

Judge each at the 75th percentile of real visits.

A worked example

A shop checks its real-user data. Desktop passes easily at 1.9 seconds. Mobile is 3.1 seconds, in the "needs improvement" band. Splitting by page template shows mobile product pages at 4.4 seconds, the worst by far: a large uncompressed product image and two blocking scripts. Compressing the image and deferring the scripts brings mobile product pages to 2.4 seconds.

How to check your own pages

1. Collect real-user timings on every page. In TapCub, page vitals are sent once when the page is hidden, and the Performance report shows P75 by default for the slowest-pages list.
2. Look at mobile and desktop separately.
3. Sort page templates by P75 LCP and start with the slowest that also get traffic.
4. Use the load timing split (backend, network, frontend) to find where time goes. More in website performance metrics.

Does load time affect bounce rate and SEO?

Slow pages lose visitors before they see anything, which shows up as fewer pageviews and lower conversion. Core Web Vitals, including LCP, are also used by Google as a page experience signal. Speed alone will not rank a page, but a slow page loses visitors whatever its position. See how to reduce bounce rate.

FAQ

What is a good page load time in seconds?

2.5 seconds or less for Largest Contentful Paint at the 75th percentile of real visits.

Is 3 seconds a good load time?

For LCP, 3 seconds is in the "needs improvement" band. It is not poor, but it is worth working on, especially on mobile.

Why is my lab test fast but real users slow?

Lab tests run on one fast device and network. Real visitors use slower phones and networks, and are farther from your server.

See also: DOMContentLoaded, Speed Index.

See also: Website optimization, Mobile CRO.