LaunchRanked

GlossarySEO

What is PageSpeed Insights?

Also called: PSI, Google PageSpeed, PageSpeed score

Definition

PageSpeed Insights (PSI) is Google's free tool that reports a page's real-user Core Web Vitals from the Chrome UX Report alongside a Lighthouse lab test with a 0–100 performance score.

PageSpeed Insights, explained

PSI shows two kinds of data. Field data at the top comes from the Chrome User Experience Report: how real Chrome users experienced the page over the trailing 28 days, with LCP, INP, CLS, FCP and TTFB. Lab data below comes from a Lighthouse run in controlled conditions, which produces the performance score and a list of diagnostics.

Field data can appear at page level or origin level. Google's PSI documentation says that when a URL doesn't have enough data, PSI falls back to origin-level data, and if the origin doesn't have enough either, no real-user data is shown. That's common for new sites. The lab score uses bands of 90 and above for good, 50 to 89 for needs improvement and below 50 for poor. Mobile tests simulate a mid-tier phone on a throttled network, which is why mobile scores are usually lower.

The two often disagree, and that's expected. Lab tests use one device, one network and an empty cache. Field data reflects your actual visitors' devices and connections. For SEO, the field data is what matters, since it's what Search Console's Core Web Vitals report is based on. Use the lab section to find causes and test fixes.

Don't chase 100. Lighthouse's own documentation says a perfect score is extremely challenging and not expected. A page that passes all three Core Web Vitals in field data is in good shape, whatever its lab score.

Why it matters for founders

PSI is the fastest free check on a page's speed. Knowing which half to trust stops you from spending a week turning a 78 into a 95 that users won't notice.

Example

PSI shows your homepage passes Core Web Vitals in field data but scores 72 in the mobile lab test. You fix the one render-blocking script it flags and move on to content.

Common mistakes

  • Treating the lab score as your ranking factor.
  • Expecting field data for a brand-new or low-traffic page.
  • Comparing mobile and desktop scores directly.
  • Chasing a perfect 100.

Sources

Checked

← All 274 glossary terms