Google Lighthouse is the best starting point for technical website performance audits, while PageSpeed Insights is better for checking how real visitors experience a page. Lighthouse runs controlled lab tests. PageSpeed Insights combines Lighthouse lab data with field data from Chrome users, when enough data exists.
TLDR: Lighthouse is an open-source audit tool that scores performance, accessibility, SEO, best practices, and web app quality. PageSpeed Insights uses Lighthouse too, but adds real-user data from the Chrome User Experience Report. For example, a retail page may score 91 in a Lighthouse lab test but show only 63% of real mobile visits passing Core Web Vitals because users have slower phones or poor networks. A serious audit should use both tools, not one or the other.
What Is Lighthouse?
Google Lighthouse is an automated website auditing tool created by Google. It tests a web page and reports how well it performs across several areas. The main categories are Performance, Accessibility, Best Practices, SEO, and Progressive Web App checks.
Lighthouse can run inside Chrome DevTools, from the command line, through Node packages, or inside other tools. It loads a page in a controlled environment, simulates throttling, records key metrics, and returns a score from 0 to 100. The score is useful, but it is not the whole story.
For many teams, it feels like Lighthouse is both helpful and irritating. A page can score 94 in one run and 88 five minutes later. That usually happens because of network variation, third-party scripts, server response time, ads, or test settings. The tool is still useful, but the numbers need context.
What Lighthouse Measures
Lighthouse performance audits focus on user-centered speed signals. These metrics help explain whether a page feels fast, stable, and usable.
- First Contentful Paint: Measures when the first visible content appears.
- Largest Contentful Paint: Measures when the main content is likely visible.
- Total Blocking Time: Shows how long scripts block user interaction.
- Cumulative Layout Shift: Measures visual jumps during loading.
- Speed Index: Estimates how quickly content appears on screen.
These metrics often point to common problems. Oversized images, render-blocking CSS, unused JavaScript, slow hosting, heavy fonts, and tag manager clutter can all hurt performance. Lighthouse lists these issues as opportunities and diagnostics, which helps developers find what to fix first.
What Is PageSpeed Insights?
PageSpeed Insights, often called PSI, is Google’s online testing tool for page performance. A person enters a URL, and PSI returns a report for mobile and desktop. The report includes Lighthouse lab results and, when available, field data from the Chrome User Experience Report.
This field data is the big difference. It reflects visits from real Chrome users over the previous 28 days. It shows how actual people experienced the page across different devices, locations, and network speeds.
That data is gold, but it can also be annoying. A team may fix a slow hero image today, then wait weeks for field data to reflect the change. PSI can also show “not enough data” for smaller sites or newer pages. Expect some waiting around, especially after large performance fixes.
Google Lighthouse vs PageSpeed Insights
Lighthouse and PageSpeed Insights are often treated as the same tool. They are closely connected, but they serve different audit needs.
| Feature | Lighthouse | PageSpeed Insights |
|---|---|---|
| Data type | Lab data | Lab data plus field data |
| Best for | Debugging and testing fixes | Checking real user experience |
| Access | Chrome DevTools, CLI, automation | Web interface and API |
| Repeatability | More controlled, but still variable | Field data changes slowly |
| Audit categories | Performance, SEO, accessibility, best practices, PWA | Mainly performance and Core Web Vitals |
Lighthouse is better for diagnosis. It tells a development team what may be slowing a page down. It can be run before and after a code change. It also works well in continuous integration, where teams can block releases that cause large performance drops.
PageSpeed Insights is better for validation. It shows whether real users are having a good experience. That matters for Core Web Vitals, search performance, and conversion rates. A perfect lab score means little if the field data shows that mobile visitors still struggle.
When to Use Lighthouse
Lighthouse works best during technical audits, redesigns, and development sprints. It gives fast feedback. A developer can compress images, rerun the test, and see the effect within minutes.
Common Lighthouse use cases include:
- Testing staging pages before launch.
- Finding unused JavaScript and CSS.
- Checking image size and format problems.
- Auditing accessibility basics such as labels, contrast, and heading structure.
- Adding performance budgets to build pipelines.
A good Lighthouse workflow uses several test runs, not one. Three to five runs give a more reliable picture. The median result is usually more useful than the best score.
When to Use PageSpeed Insights
PageSpeed Insights works best when a team needs to know how a live page performs for real users. It is useful for landing pages, product pages, articles, and any page that receives organic search traffic.
PSI is especially useful for tracking Core Web Vitals:
- Largest Contentful Paint: Should usually be under 2.5 seconds.
- Interaction to Next Paint: Should usually be under 200 milliseconds.
- Cumulative Layout Shift: Should usually stay below 0.1.
If PageSpeed Insights shows that only 58% of mobile visits pass Core Web Vitals, the page needs attention. The lab score can help identify fixes, but the field data shows how serious the issue is.
Why Scores Differ Between the Two Tools
Scores differ because the tools answer different questions. Lighthouse asks, “How does this page perform under this test setup?” PageSpeed Insights also asks, “How did real Chrome visitors experience this page recently?”
Several factors can change scores:
- Device type: Mobile tests are usually stricter than desktop tests.
- Network speed: Slow connections expose hidden performance issues.
- Server response: A busy server can delay everything.
- Third-party scripts: Ads, trackers, chat widgets, and A/B testing tools can add heavy delays.
- Location: A user far from the server may wait longer.
This is why a site can pass Lighthouse in testing but fail Core Web Vitals in the field. The lab is controlled. The real web is messy.
Best Practice for Website Performance Audits
The strongest audit process uses both tools in a clear order. First, PageSpeed Insights identifies whether real users are affected. Then Lighthouse helps diagnose the technical causes. After fixes are deployed, Lighthouse confirms immediate improvement, while PSI tracks field impact over time.
A practical audit process may look like this:
- Check PageSpeed Insights for mobile and desktop field data.
- Record Core Web Vitals and note failed metrics.
- Run Lighthouse three to five times in Chrome DevTools.
- Fix high-impact issues such as large images, blocking scripts, and slow server response.
- Retest in Lighthouse after each major fix.
- Monitor PSI field data over the next 28 days.
The goal is not a vanity score of 100. The goal is a page that loads quickly, stays stable, and responds when a visitor taps or clicks.
Which Tool Is Better?
Neither tool is better in every case. Lighthouse is the better technical audit tool. PageSpeed Insights is the better real-world performance checker. Used together, they give a fuller view of page speed and user experience.
For small sites, PageSpeed Insights may be enough for routine checks. For teams that ship code often, Lighthouse should be part of development and release testing. For ecommerce, SaaS, and publishing sites, both tools should be used regularly because small delays can affect revenue, signups, and search traffic.
FAQ
Is Google Lighthouse free?
Yes. Google Lighthouse is free and open source. It can be used in Chrome DevTools, from the command line, or through automated testing tools.
Is PageSpeed Insights the same as Lighthouse?
No. PageSpeed Insights uses Lighthouse for lab testing, but it can also show real-user field data from the Chrome User Experience Report.
Why does Lighthouse show a different score each time?
Scores can change because of network conditions, server response, background browser activity, third-party scripts, and test throttling. Multiple runs give a better average.
Which score matters more for SEO?
Core Web Vitals field data matters more than a single Lighthouse score. PageSpeed Insights is often more useful for checking that data, when enough traffic exists.
Can a page score 100 and still have problems?
Yes. A perfect Lighthouse score only reflects one controlled test. Real users may still face slow loading on older phones, weak connections, or pages filled with third-party scripts.
How often should a site be audited?
Important pages should be checked after major design, content, plugin, or script changes. High-traffic pages should be reviewed at least monthly.
