Core Web Vitals are three numbers Google uses to describe how a page feels to a real visitor. You do not need to memorise the acronyms — you need to understand three sensations and what usually ruins them.
1. Loading — Largest Contentful Paint (LCP)
What it measures: how long until the biggest thing on screen — usually a hero image or headline — is visible.
Good: under 2.5 seconds. Poor: over 4 seconds.
Usual causes: slow hosting, huge uncompressed images, no caching, render-blocking scripts, fonts loaded from another domain.
Fixes: compress and resize images, convert to WebP, enable page caching and a CDN, self-host fonts, remove unused plugins.
2. Responsiveness — Interaction to Next Paint (INP)
What it measures: how quickly the page reacts when someone taps or clicks — the delay between the tap and something visibly changing.
Good: under 200 milliseconds. Poor: over 500 milliseconds.
Usual causes: heavy JavaScript, too many third-party scripts (analytics, chat, ads), expensive page builders, sliders loading everything at once.
Fixes: remove unnecessary scripts, delay loading of non-essential ones, avoid plugins that load heavy scripts on every page, keep the number of active plugins low.
3. Visual stability — Cumulative Layout Shift (CLS)
What it measures: how much the content jumps around while loading. We have all tapped the wrong button because an image loaded above it.
Good: under 0.1. Poor: over 0.25.
Usual causes: images without width and height, ads or banners injected above content, fonts swapping after render, pop-ups appearing late.
Fixes: always set image dimensions, reserve space for ads and embeds, preload fonts or use system fonts, avoid late-loading banners at the top.
How to check your scores
- PageSpeed Insights — enter a URL, read the mobile score. Mobile matters more than desktop
- Search Console → Core Web Vitals — shows how Google sees your whole site over time
- Chrome DevTools → Lighthouse — run it locally while developing
Lab data vs field data
Lab data is a single test on a simulated device — useful for debugging. Field data is what real visitors experienced over 28 days — that is what Google uses for rankings. A great lab score with poor field data usually means real users are on slower devices or connections than your test assumed.
What actually moves the needle
- Fix your hosting — the biggest single factor
- Compress and correctly size every image
- Enable caching and a CDN
- Remove plugins and third-party scripts
- Set explicit dimensions on all media
You do not need a perfect score. You need a page that feels fast to a person on a mid-range phone — which is exactly what these three numbers describe.
