The website speed test runs Google PageSpeed Insights (Lighthouse) on your URL and shows the performance score with the five metrics that matter most: LCP, FCP, TBT, CLS and Speed Index. Test on mobile first, because Google indexes the mobile version of your site.
Key takeaways
- Mobile scores are usually lower than desktop; that is normal and it is the one to improve.
- LCP under 2.5 s and CLS under 0.1 are the targets for a good experience.
- Speed supports rankings and conversions, but content and intent still decide who ranks first.
How to run the test
- Paste the full URL of the page. Test your key landing pages, not only the home page.
- Choose Mobile or Desktop.
- Wait up to a minute while Google loads the page in a lab environment.
- Run it two or three times. Lab results can move a few points between runs.
What the metrics mean
| Metric | What it measures | Good |
|---|---|---|
| LCP | When the largest image or text block appears | ≤ 2.5 s |
| CLS | How much the layout jumps while loading | ≤ 0.1 |
| TBT | How long scripts block clicks and typing | ≤ 200 ms |
| FCP | When the first content appears | ≤ 1.8 s |
| Speed Index | How fast the visible area fills in | ≤ 3.4 s |
In the field, Google uses INP (Interaction to Next Paint) instead of TBT. TBT is its closest lab proxy: lower TBT usually means better INP.
Why speed matters for SEO and sales
Page experience is a light ranking signal, so a fast page will not outrank a much more relevant one. The bigger effect is on people. A slow page loses visitors before they read the first line, and those who leave quickly return to the results, which is the pogo-sticking pattern you want to avoid. Faster pages keep users longer and convert better, which is why we treat speed as part of conversion rate optimization (CRO), not only technical SEO.
Fixes with the biggest impact
- Images: serve WebP or AVIF, set width and height, and lazy-load images below the fold. This alone often fixes LCP and CLS.
- Hosting and cache: page cache plus a CDN cut server response time; on LiteSpeed servers, LiteSpeed Cache is a strong choice.
- Scripts: remove plugins you do not use, delay chat widgets and trackers until interaction.
- Fonts: use one or two weights and font-display: swap.
Lab data vs real users
This tool shows lab data from one simulated load. Google ranks with field data from real Chrome users over 28 days, shown in the Core Web Vitals report in Search Console. A page can pass in the lab and fail in the field, for example when real visitors use slower phones. Use the lab test to find causes, and Search Console to confirm the fix worked. If many templates fail, a technical SEO audit finds the shared cause faster than testing page by page.
Which pages to test first
Start with the pages that earn money or leads: the home page, your main service or category pages, and the landing pages you pay to send traffic to. Then test one example of each template, such as a blog post, a product page and a category page, because a fix in the template improves hundreds of URLs at once. Write down the score and the five metrics before you change anything, make one change at a time, and test again. On WordPress, the most common slowdowns we find are an oversized hero image, a page builder loading every module on every page, and three or four tracking scripts that could be loaded through a single tag manager.
Need more than a free tool?
Our team runs full audits, fixes and growth plans for sites in 10 languages.Frequently asked questions
Why is my mobile score lower than desktop?
What is a good PageSpeed score?
Why does the score change on every test?
The test says the quota is used up. What now?
Does speed directly improve rankings?
How we built this tool: the checks and thresholds follow Google Search Central documentation, Schema.org and the IETF/W3C standards, and we tune them against what we see on client sites every month. The tool runs in your browser and does not store the text or URLs you enter. Questions or a bug? Tell us.