Understand Lighthouse and Core Web Vitals

Learn how lab and field data differ and which scores count as good for LCP, INP, and CLS.

Lighthouse and Core Web Vitals both measure site performance, but they answer different questions. One shows how your page performs in a controlled test. The other shows how real visitors experience it. Using both gives you a complete picture and helps you decide what to fix first.

How they differ

Understand Lighthouse (lab data)

Lighthouse loads your page in a simulated environment — by default, a mid-tier phone on a throttled network — so every run tests the same conditions. That makes it ideal for iterating. Make a change, run Lighthouse again, and see whether the numbers moved.

Lighthouse also reports on accessibility, best practices, and search engine optimization (SEO). Those are separate categories from performance. This category focuses on the Performance score and the Core Web Vitals metrics.

Understand Core Web Vitals (field data)

Core Web Vitals are three metrics collected from real visits. Each one maps to something visitors feel:

• Largest Contentful Paint (LCP) — loading. LCP measures when the biggest visible element on the screen finishes rendering. That's usually your hero image or main headline.
• Interaction to Next Paint (INP) — responsiveness. INP measures how long the page takes to respond to clicks, taps, and key presses across the whole visit.
• Cumulative Layout Shift (CLS) — visual stability. CLS measures how much visible content moves around while the page loads.

What counts as good

Google rates each metric as good, needs improvement, or poor:

How pass or fail works

Google judges each metric at the 75th percentile (p75) of real visits, on mobile and desktop separately. A page passes when at least 75% of visits meet the "good" threshold for all three metrics. Google uses p75 so a few unusually fast or slow visits don't skew the result.

Where to find your field data

• PageSpeed Insights — the Field data section shows real-visitor LCP, INP, and CLS for a specific URL
• Google Search Console — the Core Web Vitals report groups your URLs by pass or fail status. You need to verify your site in Search Console first. Learn how to verify your site with Google.
• Chrome UX Report (CrUX) — the public dataset behind both tools

Field data only appears when a page gets enough real visits. New or low-traffic pages might show no field data yet. In that case, use Lighthouse to test your changes.

Why your results can disagree

Lighthouse simulates a mid-tier phone on a slow connection, so it often looks worse than what your laptop feels like. Your field data reflects your actual audience, which might use faster devices and networks, or slower ones. Neither report is wrong. They describe different visitors.

Here's a simple way to use both:

• Trust field data to tell you whether your pages pass
• Use lab data to find what to fix and to test your changes

Put your scores to work

• Run a Lighthouse audit and set a baseline — run your first test and record your starting scores
• Optimize your images — improve LCP by shrinking your largest images
• Prevent layout shifts — lower your CLS score
• Maintain site performance — keep scores from slipping and watch real visitor data
• Understand site performance — review what Webflow handles and what you control