How to Check Website Speed for Free (and What the Numbers Mean)
Free ways to test page speed, how to read the results, and how to decide what to fix first.
A slow website loses visitors before they see anything. Speed also feeds into Google's page experience signals through Core Web Vitals. The good news: you can measure it for free in a few minutes.
Two kinds of speed data
- Lab data โ a test run in a controlled environment on demand. Great for finding causes and testing fixes.
- Field data โ measurements from real visitors' devices over the past weeks. This is what Google uses for Core Web Vitals.
A page can look fast in a lab test on fast Wi-Fi and still be slow for real visitors on mid-range phones.
Free tools to use
- PageSpeed Insights โ shows field data (when available) and a lab test with suggestions.
- Search Console's Core Web Vitals report โ groups your pages by status using real-user data.
- Browser developer tools โ the Performance and Network panels show what loads and how long it takes.
- Multi-location test tools โ run tests from servers in different countries to see how distance affects speed.
For a quick all-round check that includes speed, mobile and technical issues, use the free site audit report.
The numbers that matter
- Largest Contentful Paint (LCP) โ when the main content appears. Good: 2.5 seconds or less.
- Interaction to Next Paint (INP) โ how quickly the page responds to taps and clicks. Good: 200 milliseconds or less.
- Cumulative Layout Shift (CLS) โ how much the layout jumps. Good: 0.1 or less.
- Time to First Byte (TTFB) โ how fast the server starts responding. Slow TTFB often means hosting or distance problems.
Read Core Web Vitals explained for beginners.
Testing for a global audience
If your server is in one country and visitors are worldwide, test from several regions. A page that loads in one second near the server may take several seconds on another continent. A CDN, caching and lighter pages reduce that gap.
What to fix first
- Oversized images โ the most common cause. See WebP vs JPG.
- Slow server response โ caching, better hosting or a CDN.
- Too many scripts โ remove unused plugins, chat widgets and trackers.
- Render-blocking CSS and fonts.
A worked example
A Canadian photography studio's homepage scored well in a lab test from North America, but visitors from Europe complained it was slow. Tests from a European location showed a server response over one second and a 4 MB hero image. Adding a CDN and compressing the hero image cut the European load time by more than half.
Related guides
More free tools: browse all our free SEO tools, no sign-up needed.
Frequently asked questions
Why are my scores different every time?
Lab tests vary with network and server conditions. Run several tests and look at the trend.
What is a good page load time?
Aim for the main content to appear within about 2.5 seconds for most visitors.
Does a perfect score matter?
No. Real-user experience matters more than a perfect lab score.
Why does my site have no field data?
Field data needs enough real visits from Chrome users. Newer or smaller sites may not have it yet.