Core Web Vitals are three metrics Google uses to measure the real-world experience of a page: LCP (loading speed), INP (responsiveness to interaction) and CLS (visual stability). You pass when, at the 75th percentile of real users, LCP is 2.5 seconds or less, INP is 200 milliseconds or less and CLS is 0.1 or less.
It sounds technical, but the logic is simple: Google wants to know whether your page loads fast, reacts when tapped and stays still while people read. The bar is real: as of May 2026, only 55.9% of sites tracked by the Chrome UX Report passed all three metrics. This guide explains each one without jargon, with the 2026 thresholds and the fixes that move them. It is part of our SEO in the AI era pillar.
What are Core Web Vitals and why do they exist?
They are the minimum set of experience metrics Google defined for every site on the web. Instead of measuring everything, it picked three dimensions: loading (LCP), responding (INP) and staying stable (CLS). The data comes from the Chrome User Experience Report (CrUX), meaning real Chrome users, not lab tests.
Grading happens at the 75th percentile over a rolling 28-day window. Translation: you pass a metric when at least 75% of real visits get a “good” experience. Your fast laptop on office wifi does not count; the average phone on mobile data does.
What are the thresholds in 2026?
The thresholds for a “good” rating remain stable:
- LCP (Largest Contentful Paint): 2.5 seconds or less. Needs improvement: 2.5 to 4s. Poor: above 4s.
- INP (Interaction to Next Paint): 200 milliseconds or less. Needs improvement: 200 to 500ms. Poor: above 500ms.
- CLS (Cumulative Layout Shift): 0.1 or less. Needs improvement: 0.1 to 0.25. Poor: above 0.25.
Pass rates show where the pain is: 68.6% of sites have good LCP, 81.3% have good CLS and 86.6% have good INP. LCP is the metric that fails most. Desktop also outperforms mobile: 57.1% versus 49.7% passing all three.
What is LCP and how do you improve it?
Largest Contentful Paint measures how long the biggest visible element takes to appear, usually the hero image or the main headline. It is the metric that best matches the feeling of “it has loaded”.
The highest-impact fixes: serve compressed, right-sized images (WebP or AVIF), prioritise the main element with fetchpriority="high", use a CDN and server caching, and remove render-blocking CSS and scripts. On most sites the LCP culprit is a heavy hero image or a slow server response.
What is INP and why did it replace FID?
Interaction to Next Paint measures the time between an interaction (click, tap, keypress) and the next visible update on screen. It replaced FID in March 2024 because FID only measured the first interaction and hid problems: a page could pass FID and still feel sluggish for the rest of the session.
INP degrades when JavaScript hogs the main thread. The fixes: reduce and split long JavaScript tasks, defer third-party scripts (analytics, chat widgets, ad pixels), and give instant visual feedback to interactions even when processing takes longer. If your site is JavaScript-heavy, INP is your biggest risk.
What is CLS and what causes it?
Cumulative Layout Shift measures how much the page “jumps” while loading. You know the experience: you go to tap a button and it moves because a banner loaded above it. CLS adds up all those unexpected shifts into one score; 0.1 or less is acceptable.
The usual causes have direct fixes: images and videos without declared dimensions (always set width and height), ads and embeds that push content down (reserve fixed space for them), and web fonts that swap late (use font-display: swap with a similar fallback). It is usually the cheapest metric to fix.
Do Core Web Vitals really affect rankings and revenue?
For rankings, they work as a tiebreaker, not a primary factor. Relevant content and authority still weigh more; between equivalent pages, the better experience wins. Strong Vitals will not save weak content, but poor Vitals lose you ties you could have won.
For the business, the effect is direct and well measured. Every extra second of LCP above 2.5 seconds cuts conversion by 4 to 8% on average. Rakuten 24 measured +53.4% revenue per visitor and +33.1% conversion rate in sessions with good LCP (web.dev case study). Vodafone improved LCP by 31% and sales rose 8%. Speed pays for itself.
How do you measure: lab data vs real users?
There are two kinds of data and you need both. Field data (CrUX) is what Google grades you on: you see the verdict in Search Console’s Core Web Vitals report and in PageSpeed Insights (the “Discover what your real users are experiencing” section). Lab data (Lighthouse, the diagnostics section of PageSpeed Insights) is for finding out why and testing fixes before shipping them.
Practical flow: use Search Console to spot failing page groups, PageSpeed Insights to diagnose one example page, ship the fix, then wait out the 28-day CrUX window to confirm. Core Web Vitals are one item in our technical SEO checklist for 2026, which covers the rest of the infrastructure layer.
Frequently asked questions
Do I need to pass all three metrics to see any benefit?
Google’s “good” status requires all three at the 75th percentile, and that is the goal. But the benefit is gradual: every improvement in LCP, INP or CLS improves experience and conversion on its own. Start with the metric where you score worst; on most sites that is LCP.
Do Core Web Vitals matter for AI answers?
Indirectly. Answer engines cite pages they can fetch and read quickly, and a slow or unstable site degrades the classic ranking signals that still feed the retrieval behind many AI answers. A fast site does not guarantee citations, but a slow one makes them harder.
How often should I check Core Web Vitals?
Monthly is enough for most sites, plus a mandatory check after any big change: redesign, new theme, new plugin, migration. Remember that CrUX uses a 28-day window: after you ship a fix, the report can take up to a month to reflect the full improvement.
