• Good thresholds: LCP 2.5 seconds or less, INP 200 milliseconds or less, CLS 0.1 or less.
• Google assesses the 75th percentile of real user data from the Chrome UX Report.
• INP replaced First Input Delay in March 2024.
• Lab tools such as Lighthouse are for diagnosis. Field data is what counts.
• It is a tiebreaker for rankings. The stronger case is usually revenue and user satisfaction.

Core Web Vitals are Google's attempt to put numbers on how a page feels to use. They feed into its page experience signals. Relevance still wins, so a slow page with the best answer will usually outrank a fast page with a poor one, but when results are close, experience can tip the balance. The commercial argument is often stronger than the SEO one: faster pages convert better.

The three metrics

Largest Contentful Paint (LCP)

How long it takes for the largest visible element, usually a hero image or main heading, to render. Good is 2.5 seconds or less. Poor is more than 4 seconds.

Interaction to Next Paint (INP)

How quickly the page responds visually after a user clicks, taps or types, measured across the whole visit. Good is 200 milliseconds or less. Poor is more than 500 milliseconds. INP replaced First Input Delay as a Core Web Vital in March 2024 and is much harder to pass because it looks at all interactions, not just the first.

Cumulative Layout Shift (CLS)

How much the layout jumps about unexpectedly. Good is 0.1 or less. Poor is more than 0.25.

How Google measures them

Google uses field data from real Chrome users, collected in the Chrome UX Report (CrUX), over a rolling 28-day window. A URL passes when 75 per cent of visits meet the good threshold for all three metrics. Where a URL lacks enough data, Google uses data from a group of similar pages or the whole origin.

Lab tools such as Lighthouse simulate a single load on a set device and connection. They are excellent for debugging and cannot measure INP properly, since nobody is interacting. Do not chase a Lighthouse score of 100. Chase passing field data.

Improving LCP

• Reduce server response time (TTFB) with caching and a CDN.
• Make the LCP image discoverable in the HTML, not injected by JavaScript or set as a CSS background.
• Do not lazy-load the LCP image. Add fetchpriority="high" to it.
• Serve modern formats (WebP or AVIF) at the right dimensions.
• Remove render-blocking CSS and JavaScript from the critical path.
• Preconnect to essential third-party origins.

 <img src="/hero.avif" width="1200" height="630" fetchpriority="high" alt="Describe the image" /> 

Improving INP

• Break up long JavaScript tasks so the main thread can respond.
• Ship less JavaScript. Audit third-party tags, which are frequently the worst offenders.
• Defer non-essential work until after the interaction has been painted.
• Keep the DOM small and avoid layout thrashing.
• Give immediate visual feedback, then do the heavy lifting.

Improving CLS

• Set width and height attributes, or a CSS aspect-ratio, on images, videos and iframes.
• Reserve space for ads, embeds and cookie banners.
• Do not insert content above existing content unless the user asked for it.
• Preload key fonts and use font-display settings and size-adjusted fallbacks to limit text shifts.
• Animate with transform instead of properties that trigger layout.

Supporting metrics

• Time to First Byte (TTFB): server responsiveness. Under 0.8 seconds is a sensible target.
• First Contentful Paint (FCP): when the first content appears.
• Total Blocking Time (TBT): a lab proxy for responsiveness problems.

Common mistakes

• Optimising for lab scores and ignoring field data.
• Testing only on a fast laptop and office connection.
• Lazy-loading everything, including the hero image.
• Adding one more tag manager script every quarter.
• Fixing the homepage and forgetting the templates that carry most of the traffic.

How to test

• Search Console Core Web Vitals report for grouped field data by template.
• PageSpeed Insights for field and lab data on one URL.
• CrUX dashboards or the API for trends and competitor comparison.
• Chrome DevTools Performance panel for debugging, including live INP interactions.
• Real user monitoring with the web-vitals library if you need more detail than CrUX gives.

The GEO angle

AI crawlers and live-fetch agents work to tight timeouts. A slow server response can mean your page is skipped in favour of one that answers quickly. Here TTFB and a lean HTML payload matter far more than visual metrics, since these systems are reading, not looking.

Add your title here

This is a paragraph. Writing in paragraphs lets visitors find what they are looking for quickly and easily. Make sure the title suits the content of this text.