What they are, and why they matter
Core Web Vitals are three real-user metrics Google uses to quantify page experience: loading (LCP), responsiveness (INP), and visual stability (CLS). They are a ranking signal — a tie-breaker between similarly relevant pages — but the bigger reason to care is conversion: slow, janky, unstable pages lose users before ranking ever enters the picture.
A URL “passes” only when all three are in the good range at the 75th percentile of real visits — i.e. the experience has to be good for at least three-quarters of your users, not just on your laptop.
The three metrics, and how to fix each
Loading — when the main content appears
- →Serve the hero image/text fast: preload it, and don’t lazy-load above-the-fold media.
- →Cut server response time (TTFB) — cache, CDN, faster origin.
- →Eliminate render-blocking CSS/JS before the main content.
- →Use modern image formats (AVIF/WebP) and correct sizing.
Responsiveness — reaction to taps/clicks
- →Break up long JavaScript tasks; yield to the main thread.
- →Defer or remove non-critical third-party scripts.
- →Debounce expensive event handlers.
- →Avoid large re-renders on interaction.
Visual stability — how much layout jumps
- →Set explicit width/height (or aspect-ratio) on images, video, and embeds.
- →Reserve space for ads, banners, and dynamically injected content.
- →Preload fonts and use font-display to avoid late layout shifts.
- →Never insert content above existing content unless in response to a user action.
Lab vs field: which number counts
This trips up almost everyone. Lab data is one synthetic test — reproducible, instant, perfect for debugging. Field data comes from the Chrome UX Report: what real Chrome users experienced over the trailing 28 days. Google ranks on field data. Use lab to diagnose; treat field as the scoreboard.
One consequence: INP can only be measured from real interactions, so lab tools (Lighthouse, PageSpeed) report Total Blocking Time (TBT) as a stand-in. A good TBT usually predicts a good INP, but only field data confirms it — which is why a serious audit reads both.
How AvocadoScore measures them
AvocadoScore pulls real-user field data from the Chrome UX Report where it exists, and falls back to a lab run (via Google PageSpeed Insights) for pages without enough field traffic — so you see the scoreboard number and the debuggable one. It does this across your whole site, not one URL, and turns the failures into a ranked fix list with effort estimates. Want a single homepage number first? The free readiness checker is instant.
Questions, answered
What are the three Core Web Vitals?+
Largest Contentful Paint (LCP) measures loading — when the main content becomes visible. Interaction to Next Paint (INP) measures responsiveness — how quickly the page reacts to taps and clicks. Cumulative Layout Shift (CLS) measures visual stability — how much the layout jumps around while loading. INP replaced First Input Delay (FID) as a Core Web Vital in March 2024.
What are the “good” thresholds?+
A URL passes when the 75th percentile of real-user visits meets: LCP at or under 2.5 seconds, INP at or under 200 milliseconds, and CLS at or under 0.1. All three must be in the “good” range at the 75th percentile for the URL to pass.
Do Core Web Vitals affect Google rankings?+
Yes, but as a tie-breaker, not a primary factor. They are part of Google’s page-experience signals. Relevance and content quality dominate ranking; Core Web Vitals help you edge out a similarly relevant competitor and, just as importantly, they directly affect conversion and bounce rate regardless of rankings.
What is the difference between lab and field data?+
Lab data is a single synthetic test in a controlled environment — reproducible, great for debugging, available instantly for any URL. Field data (from the Chrome UX Report) is what real Chrome users actually experienced over the trailing 28 days. Google ranks on field data. Lab is how you diagnose; field is the scoreboard. INP in particular can only be measured in the field, so lab tools report Total Blocking Time (TBT) as a proxy.
Why does my lab score look good but the field score is poor?+
Lab tests run once, on one device and network, often from a fast data-center connection. Real users are on slower phones, spotty networks, and cold caches, and they interact with the page in ways a single load never captures. When lab and field disagree, trust the field data — that is the population Google measures and the experience your users actually have.
See your Core Web Vitals — field and lab, whole site
AvocadoScore scores LCP, INP-proxy and CLS across your pages and hands you a ranked fix list. Free, every fix unlocked.