Core Web Vitals are Google's three user-experience metrics. They influence rankings, but more importantly they measure things your visitors feel: how fast the page shows up, how quickly it reacts, and whether it jumps around while loading.
The three metrics, in plain language
- LCP – Largest Contentful Paint (target: under 2.5 s). How long until the main content is visible.
- INP – Interaction to Next Paint (target: under 200 ms). How quickly the page responds when you tap or click.
- CLS – Cumulative Layout Shift (target: under 0.1). How much things move around unexpectedly.
Google measures these from real users' browsers (the field data in Search Console), not just lab tools. That's why a site can look fine in Lighthouse and still fail.
Our pre-launch checklist
Images & media
- Every image served in modern formats (AVIF/WebP) at the right size for the viewport
- Width and height set on all images and embeds (prevents CLS)
- Hero image preloaded with high fetch priority; everything below the fold lazy-loaded
- Videos muted, compressed, with a poster image; no autoplay on cellular
Fonts
- Self-hosted, subsetted,
font-display: swapwith metric-compatible fallbacks - At most two families, four weights
JavaScript
- Third-party scripts loaded after interaction or via a consent-aware loader
- Bundle analysed; no library imported for a single helper
- Heavy components split and loaded on demand
- Long tasks under 50 ms on a mid-range Android phone
Layout
- Reserved space for ads, embeds, cookie banners and dynamic content
- No content injected above existing content after load
- Animations use
transformandopacityonly
Delivery
- Edge caching / CDN in front of the origin
- Static or server-rendered HTML for all landing pages
- Compression (Brotli), HTTP/2 or HTTP/3
Measurement
- Lighthouse CI in the pipeline with performance budgets
- Real-user monitoring (web-vitals library → analytics)
- Search Console Core Web Vitals report checked after two weeks
The usual suspects when a site fails
- An unoptimised hero image – the number one cause of poor LCP.
- Tag-manager pile-ups – ten marketing scripts fighting for the main thread.
- Web fonts loading late – text popping in and shifting layout.
- Cookie banners injected at the top – instant CLS.
- Client-side rendering of everything – blank page until JavaScript arrives.
Does it matter for business?
Yes. Across our e-commerce projects, moving LCP from ~4 s to ~1.5 s has correlated with conversion-rate increases between 8% and 20% on mobile. Speed isn't a vanity metric – it's the cheapest conversion optimisation you'll ever do.
Want to know where your site stands? We run a free performance check for every discovery call.
PerformanceSEONext.js
