Use Google's free speed test step by step: field data vs lab data, how the score is calculated and how to turn the report into action.
If you suspect your site is slow, the first stop is usually Google's free tool, PageSpeed Insights. The tool is easy to use; the real skill is reading the report correctly, because the big score at the top does not tell you on its own how fast your site is for real users. This guide shows how to run a test and what each part of the report means.
What Is PageSpeed Insights
PageSpeed Insights is a free Google tool that measures the mobile and desktop performance of a web page. It shows two kinds of data together: field data collected from real Chrome users who visit your site, and lab data from a simulated Lighthouse test. It also gives concrete suggestions for making the page faster.
How to Run a Test
- Open pagespeed.web.dev in your browser.
- Enter the full address of the page you want to test and click analyse.
- Results appear in two tabs, mobile and desktop. Look at mobile first: many visitors arrive on a phone and Google evaluates sites by their mobile version.
Two Layers of the Report: Field Data and Lab Data
Field data: the real user experience
The top of the report shows Chrome User Experience Report (CrUX) data. It reflects the experience of real visitors over the last 28 days, assessed at the 75th percentile of visits. This is where the Core Web Vitals assessment shows as passed or failed, and it is the data Google looks at when it evaluates page experience.
If your site is new or gets little traffic, you may see a "no data" notice here. That is not an error; there are simply not enough visits to measure yet.
Lab data and the performance score
The 0-100 score further down comes from a single Lighthouse test in a controlled environment. The mobile test simulates a mid-range phone on a slow mobile connection, which is why mobile scores are lower than desktop. Read the colours like this:
- 90-100 (green): good
- 50-89 (orange): needs improvement
- 0-49 (red): poor
How the Performance Score Is Calculated
The score is a weighted average of five lab metrics:
- Total Blocking Time (TBT): 30%. How long the main thread is blocked by long tasks; the lab stand-in for responsiveness.
- Largest Contentful Paint (LCP): 25%. When the largest image or text block on the page appears.
- Cumulative Layout Shift (CLS): 25%. How much the content moves while loading.
- First Contentful Paint (FCP): 10%. When the first content appears.
- Speed Index: 10%. How quickly the page fills in visually.
We explain the three Core Web Vitals in the field data (LCP, INP and CLS) in our Core Web Vitals 2026 guide.
The Insights and Diagnostics Sections
Below the score you will find what is slowing your page down. With Lighthouse 13, released in autumn 2025, the old "opportunities" list moved to insights shared with Chrome DevTools. Common items include:
- Improve image delivery: large, old-format or oversized images
- Render-blocking requests: CSS and JavaScript that must load before the page is drawn
- LCP breakdown: which phase delays the largest content
- Cache lifetimes: files downloaded again on repeat visits
- Third parties: the cost of analytics, advertising and chat tools
- Layout shift culprits: images without dimensions, late-loading fonts
Work through the suggestions starting with the biggest estimated savings. For a step-by-step list, continue with how to increase your PageSpeed score.
Why the Score Changes Between Tests
It is normal to see different scores when you test the same page twice. The server's response time at that moment, network conditions and third-party content such as ads and analytics all affect the result. For a fair comparison, run the test three times and use the median. To see the effect of changes on real users, wait for the 28-day field data window to update.
Which Pages to Test
PageSpeed Insights measures one address at a time. Test pages that represent your different templates, not just the home page: the home page, your most visited service page, a blog post and the contact page. For field data across the whole site, use the Core Web Vitals report in Google Search Console.
Conclusion
To read PageSpeed Insights correctly, look at the field data and the Core Web Vitals assessment first, then at the lab score and the insights. A score of 100 is not the goal; the goal is real users experiencing your site as fast and smooth. If you are planning a new site, building for speed from the start is the easiest route: with our Quick Website package you see your site, built from simple mobile-friendly sections, on our test server before launch and can measure it in PageSpeed Insights yourself. For a design made for your brand, see our corporate web design service.