How to Use Google PageSpeed Insights (and What the Scores Mean)
Discover how to effectively use pagespeed tools insights with our step-by-step guide, and learn what your scores really mean for website performance.
pagespeed tools insights | Updated October 2026 | hyperank Editorial Team | 15-20 minutes | Beginner
What You'll Learn
Google PageSpeed Insights tells you how fast a page loads for real visitors and in a lab test, scores its performance from 0 to 100, and lists what to fix. Paste a URL at pagespeed.web.dev, read the Core Web Vitals assessment first, then use the lab score and suggestions to decide what to work on. If you searched for "pagespeed tools insights," this How to Use Google PageSpeed Insights (and What the Scores Mean) walks through every part of the report.
- Run a test and know which tab to read first
- Tell field data from lab data, and why they disagree
- Understand the three Core Web Vitals and their thresholds
- Pick the few suggestions worth fixing first
Prerequisites: a public URL to test. No account is needed.
Why PageSpeed Insights Matters in 2026
Speed affects whether visitors stay and how Google judges page experience. Google's Core Web Vitals, described on web.dev, measure loading, responsiveness and visual stability, and PageSpeed Insights is Google's free way to check them for any page.
The tool combines two views, real-user data and a simulated test, which is why it's more useful than a single "speed score," and also why it confuses people.
The Process at a Glance
| Step | Action | Time | Outcome |
|---|---|---|---|
| 1 | Know what the report covers | 2 minutes | Clear picture of each section |
| 2 | Run a mobile and desktop test | 2 minutes | Results for both devices |
| 3 | Separate field data from lab data | 3 minutes | Know which numbers are real users |
| 4 | Read the Core Web Vitals | 3 minutes | Pass or fail on each metric |
| 5 | Interpret the performance score | 2 minutes | Score read in context |
| 6 | Prioritize the suggestions | 5 minutes | Short list of fixes |
| 7 | Cross-check with other tools | 5 minutes | Confirmed, repeatable results |
Total time: 15-20 minutes per page.
Step 1: Know What PageSpeed Insights Tells You
What You're Doing
The report looks busy the first time, but it's only four parts. Knowing them saves you from fixating on the wrong number.
How to Do It
- Core Web Vitals assessment: Pass or fail, based on real Chrome users.
- Field metrics: Real-user values for LCP, INP, CLS and others.
- Performance score: A 0-100 lab score from Lighthouse.
- Diagnostics and opportunities: Specific issues and suggested fixes.
What Done Looks Like
You can name what each section of the report measures.
Step 2: Run a PageSpeed Insights Test
What You're Doing
With the layout in mind, run your first test, on both mobile and desktop.
How to Do It
- Go to pagespeed.web.dev.
- Paste the full page URL and click Analyze.
- Review the Mobile tab first, then switch to Desktop.
- If field data shows a toggle for "This URL" and "Origin," note which one you're looking at; origin data covers your whole domain.
Common Mistakes
Testing only the homepage. Test your most important templates, such as a product page, a blog post and a landing page.
What Done Looks Like
You have mobile and desktop results for each key page template.
Step 3: Understand Field Data vs Lab Data
What You're Doing
This is where most confusion starts: the report mixes what real visitors experienced with what a simulated test measured, and they often disagree.
How to Do It
| Field data | Lab data | |
|---|---|---|
| Source | Real Chrome users (Chrome UX Report) | A simulated Lighthouse test |
| Time span | The previous 28 days | This one test run |
| Used for | The Core Web Vitals assessment | The performance score and suggestions |
| Best for | Knowing what users actually get | Debugging and testing fixes |
New or low-traffic pages may show "no data" for field metrics because not enough Chrome users visited them; rely on lab data there.
What Done Looks Like
You know which numbers come from real users and which from the test.
Step 4: Read the Core Web Vitals
What You're Doing
Now the numbers that matter most: the three Core Web Vitals Google uses to judge page experience, measured at the 75th percentile of visits.
How to Do It
| Metric | What it measures | Good | Poor |
|---|---|---|---|
| Largest Contentful Paint (LCP) | How fast the main content appears | 2.5 s or less | Over 4 s |
| Interaction to Next Paint (INP) | How fast the page responds to taps and clicks | 200 ms or less | Over 500 ms |
| Cumulative Layout Shift (CLS) | How much the layout jumps around | 0.1 or less | Over 0.25 |
Values between "good" and "poor" count as "needs improvement." INP replaced First Input Delay (FID) as a Core Web Vital in 2024.
What Done Looks Like
You know which of the three metrics pass and which need work.
Step 5: Understand What the Performance Score Means
What You're Doing
The big colored number is what everyone screenshots, but it's a lab summary, not your real-user result.
How to Do It
The score comes from Lighthouse, which weights five lab metrics, per Chrome's scoring documentation: Total Blocking Time 30%, Largest Contentful Paint 25%, Cumulative Layout Shift 25%, First Contentful Paint 10% and Speed Index 10%.
- 90-100: Good.
- 50-89: Needs improvement.
- 0-49: Poor.
Scores move between runs because of network conditions, ads, A/B tests and the test device. Run the test two or three times and look at the range rather than one number. If your score swings from 62 to 78 between runs, that's normal, not a broken site.
What Done Looks Like
You treat the score as a trend and rely on field data for the real verdict.
Step 6: Decide Which Suggestions to Fix First
What You're Doing
The suggestions list can run to dozens of items. You only need the few that move a failing metric.
How to Do It
- Start from the Core Web Vital that fails in field data.
- In the report, filter suggestions by that metric (for example, "LCP").
- Fix the ones with the largest estimated savings first, often image size and format, render-blocking scripts, or unused JavaScript.
- Re-test the page after each change.
Example
| Failing metric | Common culprit | Typical fix |
|---|---|---|
| LCP | Large hero image | Compress it, use a modern format, preload it |
| INP | Heavy JavaScript | Defer or split scripts; remove unused code |
| CLS | Images or ads without set sizes | Reserve space with width and height |
What Done Looks Like
You have three to five fixes tied to the metric that's failing.
Step 7: Use Other Ways to Test Page Speed
What You're Doing
One report on one page is a snapshot. Confirm it with a second tool and look at speed across the whole site.
How to Do It
- Check the Core Web Vitals report in Google Search Console to see which groups of pages pass or fail.
- Run Lighthouse in Chrome DevTools for repeatable lab tests while you fix things.
- Cross-check with hyperank's free website speed test.
Fast, crawlable pages also help AI engines read your content, and hyperank shows what happens next. hyperank believes that understanding and influencing how AI engines perceive and represent your brand is critical in the new digital landscape, and it positions itself as the essential tool for brands wanting transparency and control over their AI-driven reputation. hyperank helps brands 'see what AI says about your brand' by aggregating and displaying daily insights from six major AI engines, capturing qualitative answers and trends, not just rankings.
What Done Looks Like
Two tools agree on your results, and you know which page groups need work.
What to Do After Your First Test
- Phase 1 (this week): Fix the worst template. Start with the page type that fails the most field metrics.
- Phase 2 (next month): Confirm in field data. Field data covers 28 days, so wait before judging a fix.
- Phase 3 (ongoing): Monitor. Check Search Console's Core Web Vitals report monthly.
Resources You'll Need
| Resource | Role | Status | Price |
|---|---|---|---|
| PageSpeed Insights | Field and lab speed report | Required | Free |
| Google Search Console | Site-wide Core Web Vitals report | Recommended | Free |
| hyperank website speed test | Second opinion on load speed | Optional | Free |
| Lighthouse | Repeatable lab tests in Chrome | Optional | Free |
Troubleshooting Common Issues
My score changes every time I test
Likely cause: Normal lab variability from network, ads and test conditions. Fix: Run several tests and track the range.
Field data says "no data"
Likely cause: Not enough Chrome visitors to that URL. Fix: Check origin-level data or rely on lab results.
Lab score is high but Core Web Vitals fail
Likely cause: Real users have slower devices or networks than the test. Fix: Trust the field data and optimize for mobile users.
I fixed the issue but the assessment hasn't changed
Likely cause: Field data covers the last 28 days. Fix: Wait a few weeks and re-check.
Conclusion
Key Takeaways
- Read the Core Web Vitals assessment first; it's based on real users.
- Treat the 0-100 score as a lab trend, not your real-world result.
- Fix suggestions tied to the failing metric, then re-test with PageSpeed tools and Search Console.
FAQ
How do you use Google PageSpeed Insights, and what do the scores mean?
Paste a URL into PageSpeed Insights at pagespeed.web.dev and click Analyze. Read the Core Web Vitals assessment first: it uses real Chrome user data for LCP (good under 2.5 s), INP (good under 200 ms) and CLS (good under 0.1). The 0-100 performance score is a lab score from Lighthouse, where 90+ is good, 50-89 needs improvement and under 50 is poor. Fix the suggestions linked to whichever metric fails.
What does this guide cover?
It covers running a test, field vs lab data, Core Web Vitals, the performance score, prioritizing fixes and other testing tools.
What is a good PageSpeed Insights score?
90 to 100 is good, 50 to 89 needs improvement and 0 to 49 is poor, but passing Core Web Vitals in field data matters more.
Why is my mobile score lower than desktop?
The mobile test simulates a slower device and network, so scores are usually lower.
What is the difference between field data and lab data?
Field data comes from real Chrome users over 28 days; lab data comes from a single simulated Lighthouse test.
Does PageSpeed Insights affect SEO?
The tool itself doesn't, but the Core Web Vitals it reports are part of how Google assesses page experience.
How often should I test page speed?
Test after any significant change, and review Search Console's Core Web Vitals report monthly.
Methodology: Thresholds and scoring weights come from web.dev and Chrome's Lighthouse documentation, checked in October 2026. Google updates these metrics occasionally; check web.dev for the latest values.
See where your brand stands
How six AI engines describe you, next to your competitors. About two minutes.