Core Web Vitals

The three metrics and their thresholds

• Largest Contentful Paint (LCP) — when the largest visible element, usually a hero image or headline block, finishes rendering. Good is 2.5 seconds or less.
• Interaction to Next Paint (INP) — how long the page takes to visibly respond to a tap or click, measured across the whole visit and reported near the worst case. Good is 200 milliseconds or less. INP replaced First Input Delay in March 2024, and it is stricter: FID only measured the delay before handling started, while INP measures until something actually changes on screen.
• Cumulative Layout Shift (CLS) — how much content jumps around while loading. Good is 0.1 or less.

A page passes only if all three are good at the 75th percentile of real visits, which means a quarter of your users can still have a bad experience while the page technically passes.

Field data versus lab data

This is where most confusion starts. A perfect Lighthouse score in Chrome DevTools does not mean you pass. Lighthouse is a simulated load on your machine and connection; Google's ranking use is based on the Chrome User Experience Report, which is aggregated from real Chrome users over a rolling 28-day window. Slower phones, worse networks and cold caches all show up there and not in your lab run.

The practical consequence is a delay: a fix deployed today changes the field score gradually over the following weeks.

What usually causes each failure

LCP is most often an unoptimised hero image, a render-blocking font, or a single-page application that paints nothing until its bundle executes. CLS is almost always images and ads without reserved dimensions, or a web font swapping in at a different size. INP is long JavaScript tasks blocking the main thread — heavy hydration is a frequent culprit.

Worth keeping in perspective: Core Web Vitals are a real but modest ranking signal, and mainly a tiebreaker between pages of similar relevance. Passing them will not rescue a page that does not answer the question.