Core Web Vitals explained: LCP, INP and CLS without the jargon
Google measures how a website feels to real visitors with three numbers. What LCP, INP and CLS mean, where the thresholds lie, what usually drags them down, and how much they matter for rankings.
Request an AI summary
Visitors don’t measure a website in milliseconds. They notice that a page takes a moment to show anything, that a button doesn’t react right away, or that the text jumps just as they’re about to tap a link. Core Web Vitals put numbers on exactly those three feelings.
Google uses them as part of how it judges pages, and they’re a useful health check for any site. Here’s what each one means.
1. LCP: when the main content appears
Largest Contentful Paint measures how long it takes until the largest visible element is on the screen. Usually that’s the hero image, a large heading or a video poster. It’s the moment a visitor feels “the page is here”.
- Good: 2.5 seconds or less.
- Needs improvement: up to 4 seconds.
- Poor: more than 4 seconds.
Common causes of a slow LCP:
- A huge hero image, delivered at full size to a phone, or in an old format.
- The image is found late, because it’s loaded through JavaScript or CSS instead of sitting in the HTML.
- Render-blocking files: stylesheets, fonts and scripts the browser has to load before it draws anything.
- A slow server, for example a CMS that builds every page from scratch on each request.
2. INP: how fast the page reacts
Interaction to Next Paint measures how quickly the page responds to clicks, taps and key presses. It looks at the interactions during a whole visit and reports one of the slowest. INP replaced the older metric First Input Delay in March 2024, because FID only looked at the very first interaction.
- Good: 200 milliseconds or less.
- Needs improvement: up to 500 milliseconds.
- Poor: more than 500 milliseconds.
Common causes of a poor INP:
- Too much JavaScript running on the main thread, so the browser is busy when the visitor taps.
- Third-party scripts: chat widgets, trackers, ad and consent tools, each adding work.
- Heavy handlers: a click that recalculates or redraws far more of the page than needed.
3. CLS: how much the layout jumps
Cumulative Layout Shift measures how much visible content moves around unexpectedly while the page loads and is used. It’s the reason someone taps the wrong button because an ad pushed everything down.
- Good: 0.1 or less.
- Needs improvement: up to 0.25.
- Poor: more than 0.25.
Common causes of a high CLS:
- Images and videos without fixed dimensions, so the browser doesn’t reserve space for them.
- Banners and embeds that appear above content after it’s already visible.
- Web fonts that swap in with a different size than the fallback font.
4. Field data: real visitors count
There are two kinds of measurements, and they often disagree:
- Lab data comes from a test run, such as Lighthouse in the browser’s developer tools. It’s repeatable and useful for finding causes, but it simulates one device and one connection.
- Field data comes from real visitors using Chrome, collected in the Chrome UX Report. This is what Google uses. A page passes when at least 75 percent of visits reach the “good” threshold.
A perfect lab score with poor field data usually means real visitors have slower phones or connections than the test assumed. Small sites without enough traffic may have no field data at all; then the lab data is the best guide.
5. Rankings: how much they matter
Core Web Vitals are part of Google’s page experience signals. They count, but relevance counts more: a fast page that doesn’t answer the question won’t outrank a slower one that does. Where several pages are equally relevant, speed can tip the balance.
The stronger argument is the visitors themselves. A page that appears quickly and stays steady invites people to read on, and that shows up in enquiries and sales, not just in rankings.
6. Monitoring: why once isn’t enough
Performance isn’t fixed once and done. A new plugin, a larger image uploaded by an editor, or an extra tracking script can undo it quietly. Regular checks of the field data, for example in Google Search Console, catch a decline before visitors feel it for long.
The short version
- LCP (main content visible): good at 2.5 seconds or less.
- INP (reaction to input): good at 200 milliseconds or less.
- CLS (layout stability): good at 0.1 or less.
- Field data from real visitors is what counts; 75 percent of visits need to be “good”.
- Rankings: relevance first, speed as a tie-breaker; visitors notice it either way.
If you’d like to know how your site does on Core Web Vitals and why, get in touch.