Skip to main content
Free Tool

Core Web Vitals Explainer

Enter your LCP, CLS and INP scores — see how Google grades each one and get a prioritised fix list for whichever metric is failing hardest. Thresholds updated to March 2024's INP rollout.

LCP + CLS + INPGood / NI / Poor bandsPer-metric fix listOverall pass/failGoogle 2024 thresholdsIn-browser only

Your scores

1

Enter your Core Web Vitals

Get real-user values from Google Search Console → Core Web Vitals, or PageSpeed Insights (field data section).

Core Web Vitals assessment

OverallFail — at least one metric needs improvement
LCPLargest Contentful Paint
Needs Improvement
3.2s
02.5s4s+

How long until the biggest visible element renders. Measures perceived load speed — the moment users feel the page has loaded.

Top fixes
  • Preload the LCP image with <link rel="preload" as="image">
  • Add fetchpriority="high" to the hero image
  • Serve AVIF/WebP with responsive srcset
CLSCumulative Layout Shift
Needs Improvement
0.14
00.10.25+

How much the page visually shifts while loading. Measures visual stability — no one wants to tap a button that just jumped.

Top fixes
  • Set explicit width and height on every <img>
  • Reserve space for ads and embeds with aspect-ratio
  • Avoid injecting DOM above existing content
INPInteraction to Next Paint
Needs Improvement
280ms
0200ms500ms+

Responsiveness to real user interactions (clicks, taps, keypresses). Replaced FID in March 2024 — more realistic measure of perceived snappiness.

Top fixes
  • Break up long JavaScript tasks (over 50ms)
  • Defer non-critical scripts and third-party tags
  • Use requestIdleCallback for heavy work
Summary
  • warningCore Web Vitals assessment: Fail (needs improvement). Fix the metric in the red band first.
  • warningLCP is 3.2s — Needs Improvement. Target: ≤ 2.5s.
  • warningCLS is 0.14 — Needs Improvement. Target: ≤ 0.1.
  • warningINP is 280ms — Needs Improvement. Target: ≤ 200ms.

How Google Actually Measures Core Web Vitals

Four things every team gets wrong about CWV scoring.

1

Field data, not lab data

Google uses Chrome User Experience (CrUX) field data from real users — not Lighthouse lab runs. A page can score 100 in Lighthouse and still fail Core Web Vitals because actual users hit slower conditions.

2

75th percentile, not average

Google uses p75 — 75% of page views must be in the Good band. If 74% are fast and 26% are slow, you still fail. Focus on the slow tail, not the median.

3

All three must pass

LCP Good + CLS Good + INP Needs Improvement = overall Fail. You don't get credit for two out of three. The weakest link determines the grade.

4

Separate desktop and mobile

CWV is measured separately for each device class. Mobile is usually the harder bar — plan for mobile first, desktop second. Check both in Search Console.

Core Web Vitals Threshold Reference

Exact Google thresholds, updated to the March 2024 INP rollout.

LCP — Good

≤ 2.5 seconds. The largest visible content (image, video, block of text) must paint within 2.5s at the 75th percentile.

LCP — Poor

> 4 seconds. Over 4s LCP means users see a blank or skeletal page long enough to notice. Bounce rate climbs sharply.

CLS — Good

≤ 0.1. Almost no layout shift. Feels stable. Achievable with explicit dimensions on images, ads, and embeds.

CLS — Poor

> 0.25. Heavy visual instability — content jumps, users mis-click. Usually caused by late-loading banners, ads, or unset image dimensions.

INP — Good

≤ 200ms. Replaces FID since March 2024. Measures the 98th-percentile interaction response across a visit, not just the first.

INP — Poor

> 500ms. Users feel the app is broken. Usually caused by heavy JavaScript on the main thread during interactions.

TTFB (bonus)

Not a CWV metric but strongly correlated with LCP. Aim < 600ms from server to first byte — anything over 1s often drags LCP into Poor.

FID (retired)

First Input Delay was the old interactivity metric. Retired March 2024 — replaced by INP which captures all interactions, not just the first.

What Actually Breaks Each Metric

The recurring culprits we catch in performance audits.

LCP killers

Hero image lazy-loadedSlow TTFBRender-blocking CSS/JSUnoptimised hero (no WebP)No preload hintFont-swap delay on LCP text

CLS triggers

Images without dimensionsLate-loading adsBanner injected above foldCookie bar pushing contentWeb fonts swapping lateLazy-loaded iframes

INP villains

Long JS tasks (>50ms)Heavy event handlersLayout thrashingThird-party ad SDKsReact hydration mismatchDebounce/throttle missing

Quick wins

fetchpriority="high" on herowidth + height on <img>async/defer on scriptsaspect-ratio for iframesfont-display: swapcode-split heavy routes

Core Web Vitals FAQ

What score do I need to pass Core Web Vitals?

All three metrics must be in the "Good" range for 75% of page views: LCP ≤ 2.5s, CLS ≤ 0.1, INP ≤ 200ms. Failing even one bucket means you don't pass the overall Core Web Vitals assessment.

Where do I get my real scores?

Field data (real users) from Google Search Console, CrUX Dashboard, or PageSpeed Insights. Lab data (synthetic) from Lighthouse or WebPageTest. Google uses field data for ranking, not lab.

Is CWV a ranking factor?

Yes, since 2021 (Page Experience update). It's a tie-breaker rather than a dominant signal — great content with poor CWV can still rank well, but on equal content, CWV is the differentiator.

Is my input sent anywhere?

No. Everything runs locally in your browser. Nothing is uploaded or stored.

Want Your Core Web Vitals Fixed, Not Just Explained?

Our Australian performance team diagnoses LCP, CLS and INP issues on your site, ships the fixes, and sets up CI regression checks so the scores stay Good.

  • LCP / CLS / INP diagnostics
  • Deploy-ready fix patches
  • No lock-in commitment
Book a Free Consultation First
🔒 Secure checkout|Delivered within 48 hours|100% money-back guarantee

No long-term commitment. Cancel anytime. 100% satisfaction guaranteed.