Core Web Vitals (CWV)

Core Web Vitals (CWV) is the user-experience metric set Google announced in 2020 and folded into ranking. The bigger contribution: it turned UX from gut-feel into something you can actually measure. Three axes, separate scores for mobile and desktop, and the headline numbers come from real-user data (CrUX, 28-day rolling) — which is why your Lighthouse lab score and your Search Console field score will not match.

The specific thresholds are worth memorising. **LCP (Largest Contentful Paint)** measures when the biggest visible element finishes rendering — under 2.5s is `Good`, over 4.0s is `Poor`. **INP (Interaction to Next Paint)** replaced FID in March 2024 and tracks how long it takes for the next frame to paint after a user interaction — under 200ms is `Good`, over 500ms is `Poor`. **CLS (Cumulative Layout Shift)** sums unexpected layout movement during load — under 0.1 is `Good`, over 0.25 is `Poor`. All three need to land in `Good` for the page itself to count as `Good`.

In practice CWV is a tie-breaker, not a knockout punch. Between two pages of similar quality, the one with healthier vitals gets a small edge. Trying to rescue weak content with vitals work almost always fails. The flip side: solid content stuck in `Poor` is clearly being penalised and is the highest-ROI fix. Common levers are image lazy-loading and compression, font preloading, fixed-size ad slots, deferring heavy third-party scripts (chat, heatmaps) and JavaScript bundle splitting.

The real reason to do this work, though, sits outside SEO: the same fixes lift conversion rate directly. Studies from Google, Deloitte and Amazon repeatedly show that taking 1s off mobile LCP correlates with 5–20% conversion lift. For ecommerce, the conversion effect often dwarfs the SEO effect, which means CWV should be prioritised as revenue work, not SEO work. In Korea, where many ecommerce stores see 80%+ mobile traffic share, CWV functions as a core mobile KPI in practice.

For measurement, combine PageSpeed Insights (field + lab data), Search Console's Page Experience report (sitewide trends) and Chrome DevTools Performance panel (single-page debugging). CrUX needs enough traffic to surface a score, so brand-new sites lean on Lighthouse lab scores as a stand-in until field data populates.