Core Web Vitals: Diagnose LCP, INP, and CLS

Updated September 20, 2026· 7 min read

Core Web Vitals are Google’s field metrics for loading, interaction responsiveness, and visual stability. The current set is Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS). web.dev’s Web Vitals article (updated October 31, 2024) and Google Search’s Core Web Vitals page use the same good thresholds: LCP within 2.5 seconds, INP at 200 milliseconds or less, and CLS at 0.1 or less.

Passing those thresholds is a user-experience target, not a ranking promise. Google’s page-experience guidance says ranking systems use Core Web Vitals, and that good Search Console or third-party scores do not guarantee a top result. Relevance still wins when pages compete. Web design and development work should still chase the field metrics because visitors feel them.

Thresholds, field data, and lab data

Google classifies a page as passing Core Web Vitals when all three metrics meet the good target at the 75th percentile of real page loads, split by mobile and desktop. The 75th percentile is the value that 75 percent of visits meet or beat. It is not an average.

Metric	What it measures	Good	Needs improvement	Poor
LCP	Time until the largest visible image or text block renders	2.5 s or less	Up to 4 s	Over 4 s
INP	Time from click, tap, or key press until the next paint	200 ms or less	Up to 500 ms	Over 500 ms
CLS	Unexpected layout movement during the page’s life	0.1 or less	Up to 0.25	Over 0.25

Search Console’s Core Web Vitals report uses Chrome User Experience Report (CrUX) field data over a rolling 28-day window. URLs are grouped by similar templates. Once a group has enough LCP and CLS data, the group’s status is its worst metric. Mobile and desktop are separate tabs. Low-traffic URLs can show no row.

Field versus lab. Field data is what real Chrome users experienced. Lab data is a simulated load (Lighthouse, a local DevTools trace, WebPageTest). Lab tools are the right place to reproduce a cause before release. They are not a substitute for CrUX. A standard load-only Lighthouse run does not observe the real interactions needed for field INP, so it reports Total Blocking Time as a diagnostic proxy.

INP replaced FID in March 2024. FID is historical.

Symptom-to-cause table

Symptom in field data	Likely lab trace	Common root cause	First repair to try
LCP is slow, TTFB is also slow	Document request waits	Server, redirects, cache miss, distant origin	Reduce TTFB: fewer redirects, cache HTML, closer origin
LCP is slow, TTFB is fine, FCP is late	Render-blocking CSS or JS	Stylesheets or scripts delay first paint	Inline critical CSS, defer non-critical JS
LCP is a large image that starts late	Resource load delay	Hero image not in HTML, lazy-loaded, or low priority	Discover the image from HTML; do not lazy-load the LCP image
LCP image loads, then still waits	Element render delay	Client-side render, main-thread work, or waiting on other work	Render the LCP element as soon as its resource is ready
INP is poor on a template with menus, filters, or carts	Long tasks around the interaction	Input delay, heavy handlers, or presentation delay	Break long tasks; show immediate visual feedback
CLS is poor on article or PDP templates	Shifts after load	Unsized images, ads, embeds, webfonts, injected banners	Reserve space; avoid inserting content above existing content

web.dev’s LCP optimization guide splits LCP into four subparts that add up to the whole metric: Time to First Byte, resource load delay, resource load duration, and element render delay. Delay subparts should stay small. INP observes click, tap, and keyboard interactions, not hover or plain scrolling. CLS counts unexpected shifts. Shifts within 500 milliseconds of discrete input are generally excluded.

Third-party tags, chat widgets, and ads show up in all three metrics. Treat them as first-class suspects when a template fails in the field but looks fast on a quiet local profile.

Diagnosis workflow

1. Read field status. In Search Console, note device, URL group, which metric is Poor or Needs improvement, and the 75th-percentile value. If there is no data, the URL group may be below CrUX volume. Use origin-level CrUX or your own RUM instead of guessing.
2. Pick a page template. Open a representative URL from the group, not a one-off campaign landing page unless that URL is the group.
3. Compare field and lab. PageSpeed Insights shows CrUX (if available) and a Lighthouse lab run. If field LCP is poor and lab LCP is good, reproduce a slower device or network. If both are poor, the lab trace is enough to start.
4. Trace the cause. In Chrome DevTools Performance, identify the LCP element and its subparts, the INP interaction (or TBT long tasks in Lighthouse), or the layout-shift sources.
5. Fix one cause, then validate. Confirm the lab trace improved. Ship. Field reports lag because Search Console uses 28 days of CrUX, so a single Lighthouse score is a lab checkpoint rather than production proof.

This measurement work sits next to how the page appears on a SERP. Speed is not an E-E-A-T factor.

Frequently Asked Questions

What are the three Core Web Vitals?

Largest Contentful Paint (LCP) for loading, Interaction to Next Paint (INP) for responsiveness, and Cumulative Layout Shift (CLS) for visual stability. Google treats all three as stable Core Web Vitals.

Did INP replace FID?

Yes. INP replaced First Input Delay as the responsiveness Core Web Vital in March 2024 because FID only measured delay on the first input. INP looks at click, tap, and key interactions across the visit.

What counts as a good Core Web Vitals score?

At the 75th percentile of real visits, good means LCP at 2.5 seconds or less, INP at 200 milliseconds or less, and CLS at 0.1 or less. Search Console also uses Needs improvement and Poor bands above those values.

Are Core Web Vitals a Google ranking factor?

Google says Core Web Vitals are used by ranking systems as part of page experience. Good report scores do not guarantee top rankings. Relevance and content quality still decide which page Search shows.

How should a team check Core Web Vitals?

Start with Search Console’s field report for URL groups, then open a representative URL in PageSpeed Insights. Use a lab trace to find the cause. A standard load-only Lighthouse run cannot measure field INP, so Total Blocking Time is a diagnostic proxy.

Sources

• web.dev, Web Vitals, updated October 31, 2024, accessed September 20, 2026.
• web.dev, Optimize Largest Contentful Paint, updated March 31, 2025, accessed September 20, 2026.
• web.dev, Interaction to Next Paint, updated September 2, 2025, accessed September 20, 2026.
• web.dev, Cumulative Layout Shift, accessed September 20, 2026.
• Google Search Central, Understanding Core Web Vitals and Google search results, accessed September 20, 2026.
• Google Search Central, Understanding page experience in Google Search results, accessed September 20, 2026.
• Search Console Help, Core Web Vitals report, accessed September 20, 2026.
• PageSpeed Insights, About, accessed September 20, 2026.
• Google Search Central blog, Introducing INP to Core Web Vitals, May 10, 2023.