Free Lighthouse website speed test
Is your website fast enough? Test any public URL with Google Lighthouse. See your performance score, loading measurements and the fixes likely to make the biggest difference.
Understand your test
MetaLens runs a fresh Google Lighthouse lab test, then turns the measurements into plain-English priorities. It does not treat one score as a promise about rankings or every visitor's experience.
Choose mobile or desktop and enter a public page. Google's PageSpeed Insights service loads it under controlled Lighthouse conditions, measures the page and returns diagnostic opportunities. Tests usually take 30–90 seconds.
Lighthouse is a simulated test that helps diagnose a page now. Core Web Vitals use aggregated real-user experience over time. This report currently shows Lighthouse lab measurements, not a field Core Web Vitals assessment.
Scores from 90–100 are considered good, 50–89 need improvement and below 50 is poor. Use the score to find work, then focus on the measurements and recommendations behind it.
Six lab measurements
How long the largest visible image or text block takes to appear.
How long heavy main-thread work blocks the page during loading. TBT is a useful lab diagnostic, not field INP.
How much visible content unexpectedly shifts while the page loads.
How quickly the first text, image or other visible content appears.
How quickly the visible parts of the page become complete.
How long the server takes to begin returning the page.
Why results differ
Mobile Lighthouse testing simulates a more constrained device and network, so its score is often lower than desktop. Hosting conditions, third-party services and test variability can also change results between runs.
Common causes
Transparent methodology
Tests use Google's PageSpeed Insights API and Lighthouse. Results are simulated lab measurements for the selected device and can vary between runs. MetaLens currently does not present these results as real-user Core Web Vitals.
Useful answers
Pages and test environments are not perfectly static. Server load, third-party requests, network conditions and the content returned during a run can all affect the result. Look for repeated patterns rather than overreacting to one point.
No. Performance contributes to user experience, but a Lighthouse score is not a ranking guarantee. Useful content, relevance, accessibility and many other signals still matter.
No. TBT is a Lighthouse lab diagnostic. The real-user Core Web Vital for responsiveness is INP. Improving heavy JavaScript that hurts TBT may also help INP, but they are not the same measurement.
Start with mobile for most public websites because it represents the more constrained experience. Test desktop separately when it matters to your audience, then fix issues that repeatedly appear in either report.