Core Web Vitals are three metrics Google uses to measure real world page experience: LCP, INP and CLS. The targets are 2.5 seconds, 200 milliseconds and 0.1.
Core Web Vitals are three metrics Google uses to measure real world page experience: Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift. The targets are LCP under 2.5 seconds, INP under 200 milliseconds and CLS under 0.1. You pass only if all three are met for 75% of visits.
Largest Contentful Paint
LCP measures how long it takes for the biggest element in the viewport to render, usually a hero image, a video poster or a large block of text. Under 2.5 seconds is good, 2.5 to 4 seconds needs improvement, over 4 seconds is poor.
Most failures come from three places: a slow server response before anything can start, render blocking CSS and JavaScript in the head, and oversized images that were never compressed. Preloading the hero image and serving it in a modern format usually moves the number more than anything else you can do in an afternoon.
Interaction to Next Paint
INP measures responsiveness. It watches every click, tap and key press during a visit and reports close to the worst one: the time from the interaction to the next frame the browser paints. Under 200ms is good, 200 to 500ms needs improvement, over 500ms is poor.
INP replaced First Input Delay in March 2024. FID only measured the delay before the browser started processing the first interaction, which flattered plenty of sites that felt sluggish to actually use. INP is harder to pass and a much better proxy for how a page feels. Heavy third party tags, oversized JavaScript bundles and expensive event handlers are the usual culprits.
Cumulative Layout Shift
CLS measures visual stability, meaning how much content jumps around while the page loads. Under 0.1 is good, 0.1 to 0.25 needs improvement, over 0.25 is poor. Images and iframes with no width and height attributes, ads injected into the content flow, cookie banners and web fonts that swap in at a different size cause nearly all of it.
Field data beats lab data
Google scores you on field data from the Chrome User Experience Report: anonymised measurements from real Chrome users, aggregated over a rolling 28 day window and reported at the 75th percentile. A pass means three quarters of visits hit the threshold, not that the average visit did.
Lab tools such as Lighthouse simulate one load on a throttled connection. They are good for diagnosis and useless as a scorecard. A Lighthouse score of 98 alongside failing field data just means your test conditions were kinder than your visitors’ phones. Lighthouse cannot measure INP at all, because there is no real interaction to time. Read the Core Web Vitals report in Search Console for the number that counts, then use lab tools to work out why it looks like that.
How much they affect rankings
Less than the coverage suggests. Core Web Vitals sit inside Google’s page experience signals, and Google has been consistent that they act as a tiebreaker between pages of similar relevance. A slow page with the best answer generally still wins. The stronger argument is commercial: slow, jumpy pages lose conversions, and that shows up in revenue long before it shows up in rankings. We treat them as a technical SEO fix with a direct payback rather than a ranking play.
Related reading: our guides to page speed, how to improve Core Web Vitals, Cumulative Layout Shift and Largest Contentful Paint.
Common questions
Are mobile and desktop scored separately?
Yes, they are separate datasets and you can pass one and fail the other. Mobile is the one to prioritise, both because of mobile first indexing and because that is where the slow devices and connections are.
How long after a fix before the numbers move?
Up to 28 days, because the field data is a rolling 28 day window. You will see partial movement sooner as new sessions dilute the old ones. Deploy, then leave it a month before you judge the result.
What happened to First Input Delay?
Google retired it in March 2024 and replaced it with INP. Most sites that comfortably passed FID do not automatically pass INP, so it is worth rechecking if nobody has looked since then.
For the short version and the rest of the performance vocabulary, see Core Web Vitals in our glossary.
