The metrics Google measures — and so do we

Core Web Vitals are three metrics Google uses to measure how a site feels to a real user. LCP — Largest Contentful Paint — measures how long until the main content of a page becomes visible. INP — Interaction to Next Paint — measures how quickly the page reacts to a tap or click. CLS — Cumulative Layout Shift — measures how much the layout unexpectedly jumps while loading. Together they're a reasonable approximation of 'does this site feel fast and stable'.

Most agency-built sites fail at least one of these, and CLS is the most common culprit. The fix is usually simple but easy to skip during development: images without explicit width and height attributes shift the layout when they load. Web fonts that aren't preloaded cause text to reflow. Third-party ad or chat widgets that insert themselves into the DOM push content down. None of this is visible on a developer's fast machine — it only shows up on slower devices or connections.

We measure from day one. Every project gets Lighthouse checks in CI, so a pull request that regresses LCP or introduces CLS fails before it's even reviewed. We also use real-user monitoring through Vercel Analytics to track field data — what actual users experience, not just what the test machine shows. This creates accountability at the code level rather than a panicked post-launch audit.

The business case for this is direct. Google uses Core Web Vitals as a ranking signal in search results — a poor score is a concrete SEO cost. But beyond rankings, faster and more stable pages simply convert better. Users who don't encounter layout shifts are more likely to click. Pages that respond instantly feel trustworthy. These aren't abstract improvements; they're metrics that map to revenue.