Free SEO audit for new clients — claim yours
Home/Blog/Technical SEO/How to Use PageSpeed Insights
Technical SEO

How to Use PageSpeed Insights

How to Use PageSpeed Insights
Table of contents
  1. Key Takeaways
  2. The two halves of a PSI report
  3. Reading the Core Web Vitals assessment
    1. “This URL” versus “Origin”
  4. Understanding the performance score
    1. Why the score changes between runs
  5. Insights and Diagnostics
  6. How to use PageSpeed Insights: step by step
  7. Common mistakes
  8. Related Guides
  9. Frequently Asked Questions
    1. Is a PageSpeed Insights score of 90 or more required for SEO?
    2. Why does my PageSpeed Insights score change every time?
    3. Why is there no field data for my page?
    4. What is the difference between PageSpeed Insights and Lighthouse?
    5. How long does it take for fixes to show in PageSpeed Insights?
  10. Conclusion
  11. References

PageSpeed Insights (PSI) is the free Google tool most people use to check how fast a page is. Paste in a URL, wait a few seconds, and you get a colourful score out of 100, a Core Web Vitals assessment and a long list of suggestions. It is genuinely useful, and also widely misread.

The most common mistake is treating the big performance score as the result that matters. It is not. The score comes from a single simulated test, while Google’s Core Web Vitals assessment comes from real Chrome users visiting your site over the previous 28 days. The two can disagree, sometimes dramatically, and knowing why is the key to using the tool well.

This guide walks through every section of a PageSpeed Insights report in 2026, explains field versus lab data, how the score is calculated and why it changes between runs, and gives a repeatable process for turning the report into fixes that improve real-world performance.

Key Takeaways

  • PSI shows two kinds of data: field data from real users (Chrome UX Report) and lab data from a single Lighthouse test.
  • The Core Web Vitals assessment at the top uses field data and is what aligns with Google’s page experience evaluation.
  • The 0–100 performance score is lab-only, weighted from five metrics, and naturally varies between runs.
  • When a URL lacks enough real-user data, PSI may show origin-level data for the whole site instead.
  • Use the Insights and Diagnostics sections to find causes, then confirm fixes in field data over about four weeks.

The two halves of a PSI report

Field data (“Discover what your real users are experiencing”)Lab data (“Diagnose performance issues”)
SourceChrome User Experience Report (CrUX)Lighthouse running a simulated page load
WhoReal, opted-in Chrome usersOne emulated device and network from a Google data centre
Time periodTrailing 28 daysRight now, one load
MetricsLCP, INP, CLS, plus FCP and TTFBFCP, LCP, TBT, CLS, Speed Index
Best forKnowing whether real users have a good experienceDebugging and testing changes quickly
Measures interactions?Yes (INP)No; Total Blocking Time is a proxy

“Lab data is useful for debugging issues, as it is collected in a controlled environment.”

— Google for Developers, About PageSpeed Insights

Reading the Core Web Vitals assessment

The field data section opens with a pass or fail verdict. Google’s PSI documentation explains the rules: when there is enough data for INP, LCP and CLS, the page passes only if the 75th percentile of all three is “good”. If INP has insufficient data, the page can pass on LCP and CLS alone. If LCP or CLS lacks data, it cannot be assessed.

Below the verdict you will see a distribution bar for each metric (the share of visits rated good, needs improvement and poor) and the 75th percentile value. The thresholds are the standard ones covered in our Core Web Vitals guide: LCP 2.5 seconds, INP 200 milliseconds and CLS 0.1.

“This URL” versus “Origin”

CrUX only includes pages and origins that are publicly discoverable and have enough real-user visits. If a specific URL does not meet the threshold, PSI falls back to origin data, which covers your whole site. That is useful context but does not tell you about the specific page. In that case, rely on the Search Console Core Web Vitals report, which groups similar URLs, or collect your own field data with the web-vitals library.

Note also that CrUX comes only from Chrome on supported platforms; Chrome on iOS, for instance, does not contribute. Field data is a large sample of real experiences, not a census.

Understanding the performance score

The score at the top of the lab section is calculated by Lighthouse from five metrics, weighted as follows in current versions:

Lab metricWeightWhat it reflects
Total Blocking Time (TBT)30%Main-thread blocking during load (a proxy for responsiveness)
Largest Contentful Paint (LCP)25%When the main content appears
Cumulative Layout Shift (CLS)25%Visual stability during load
First Contentful Paint (FCP)10%When anything first appears
Speed Index10%How quickly the visible page fills in

Scores of 90–100 are rated good (green), 50–89 needs improvement (orange) and 0–49 poor (red). A green score is a useful goal, but PSI’s own documentation notes that a good lab score does not guarantee a good real-user experience.

Why the score changes between runs

“A lot of the variability in your overall Performance score and metric values is not due to Lighthouse.”

— Chrome for Developers, Lighthouse performance scoring

Chrome lists causes such as A/B tests or changing ads, internet traffic routing, different test hardware, browser extensions and antivirus software. In PSI specifically, network and server conditions at the moment of the test matter. Run three to five tests and look at the median rather than reacting to one result.

Insights and Diagnostics

Below the metrics, PSI lists opportunities to improve. In 2025, Chrome replaced many of Lighthouse’s individual performance audits with consolidated “insights” that match the Chrome DevTools Performance panel. You will now see items such as:

  • LCP breakdown and LCP request discovery: which element is the LCP and where its time goes. See our LCP guide.
  • Render blocking requests: CSS and scripts delaying the first paint. See how to eliminate render-blocking resources.
  • Document request latency: redirects, slow server response and missing compression. See Time to First Byte.
  • Improve image delivery: oversized, poorly compressed or old-format images. See image compression.
  • Layout shift culprits: elements causing CLS.
  • Third parties, Use efficient cache lifetimes and Optimize DOM size.

Estimated savings are useful for prioritising but are estimates; fixing an item with “1.2 s” savings rarely cuts exactly 1.2 seconds from field LCP.

How to use PageSpeed Insights: step by step

  1. Test the right pages. Pick one representative URL per template (home, category, product or service, article, landing page) rather than only the homepage.
  2. Start with field data on mobile. Note pass or fail, which metric fails, and whether the data is URL-level or origin-level.
  3. Cross-check Search Console. The Core Web Vitals report shows how many URLs in each group are affected; see our Search Console guide.
  4. Run the lab test three to five times and note the median score and LCP element.
  5. Map insights to the failing field metric. If field LCP fails, prioritise LCP-related insights; if INP fails, look at TBT, third parties and JavaScript.
  6. Fix by template, so one change improves hundreds of URLs.
  7. Re-test the lab immediately, then monitor field data for about four weeks as the 28-day window rolls forward.
  8. Automate monitoring with the PageSpeed Insights API or a scheduled Lighthouse audit for key URLs.

Common mistakes

  • Chasing 100. A perfect lab score is not required to pass Core Web Vitals, and the last few points rarely matter to users.
  • Ignoring field data because the lab score looks fine, or panicking about a lab score when field data passes.
  • Testing only the homepage. Product and article templates often perform very differently.
  • Comparing desktop and mobile scores directly. Mobile tests use a slower simulated device and network, so they are almost always lower.
  • Treating every suggestion as mandatory. Prioritise insights linked to your failing metrics.
  • Expecting instant field changes. Real-user data takes weeks to reflect a fix.

Remember that Google describes good Core Web Vitals as aligning with what its core ranking systems seek to reward, but they are one signal among many. Speed work pays off most when it also improves conversions; our page speed guide covers the wider picture.

Frequently Asked Questions

Is a PageSpeed Insights score of 90 or more required for SEO?

No. Google uses field Core Web Vitals data, not the lab score. A page can pass Core Web Vitals with a lab score below 90, and vice versa.

Why does my PageSpeed Insights score change every time?

Lab tests are single simulated loads affected by network conditions, server load, ads and third-party scripts. Run several tests and use the median.

Why is there no field data for my page?

CrUX only includes public pages and origins with enough visits from opted-in Chrome users. Low-traffic pages may show origin data or none at all.

What is the difference between PageSpeed Insights and Lighthouse?

PSI runs Lighthouse for its lab data and adds real-user field data from the Chrome UX Report. Lighthouse alone only provides lab data.

How long does it take for fixes to show in PageSpeed Insights?

Lab results change immediately. Field data covers a trailing 28-day period, so improvements appear gradually over about four weeks.

Conclusion

PageSpeed Insights is most valuable when you read it in the right order: field data first to see what real users experience, lab data and insights second to find out why. Test each template, fix the issues tied to failing metrics and give field data time to update. If you would like an expert to turn your PSI reports into a prioritised fix list, our page speed optimization service can help, or get a free quote.

References

  1. About PageSpeed Insights – Google for Developers
  2. Lighthouse performance scoring – Chrome for Developers
  3. Moving Lighthouse to performance insights – Chrome for Developers
  4. CrUX methodology – Chrome for Developers
  5. Understanding Core Web Vitals and Google search results – Google Search Central
  6. Google PageSpeed Insights: What It Is and How to Boost Your Score – Semrush Blog
Rohit Choudhary
Written byRohit ChoudharyCo-Founder, My SEO Experts

Rohit Choudhary is the co-founder of My SEO Experts and has more than 14 years of hands-on experience in SEO, paid advertising and digital marketing. Since starting out in 2011 he has helped hundreds of businesses grow through practical, data-driven search strategies, and he also serves as a consulting Digital Marketing Director for several US-based companies.

Keep reading

Related articles

Ready to grow your business online?

Get a free, no-obligation proposal tailored to your goals and budget.

CallFree Quote