Originally published on CodingCush.
You can check your website speed for free with Google PageSpeed Insights. Go to the tool, paste your page address, click Analyze, and within a few seconds Google gives you scores out of 100 for mobile and desktop, plus a clear list of what is slowing your page down and how to fix it. You do not need to install anything, create an account, or understand code to run your first test. This guide walks you through the whole report in simple words, so the numbers actually make sense.
How to Run the Test, Step by Step
Open Google PageSpeed Insights in your browser. You will see one simple box asking you to enter a web page URL. If you are not fully sure what a URL is, our plain guide on what a URL is explains it in two minutes. Type or paste the full address of the page you want to test, including the https part, then click the Analyze button.
Wait for around 30 seconds while Google loads your page on a simulated phone and a simulated computer. Do not close the tab, even if it feels slow. That wait is the tool doing real work, not the tool being lazy.
When the report opens, look at the top first. You will see two tabs, Mobile and Desktop. Always check the Mobile tab first, because most visitors use phones and Google mainly ranks your site using the mobile version. Note your Performance score, check whether you passed the Core Web Vitals assessment, and only then scroll down to the detailed suggestions.
Test more than your homepage. Run the test on a blog post, a product page, and your contact page too, because one heavy page can be slow while the rest of the site is fine. Run each test two or three times and take the middle result, since scores can jump by a few points between runs. Keep a simple note of the date, the page, and the score each time you test, so you can see real progress over weeks instead of guessing from memory. It takes less than a minute.
What the 4 Scores Mean
The report gives four scores, each out of 100. Scores from 90 to 100 are green and good, 50 to 89 are orange and need work, and below 50 is red and poor. Here is what each one actually measures.
Performance is the speed score, and it is the one most people care about. It is built from timings such as how fast the first content appears and how quickly the page becomes usable. A score below 50 means visitors are waiting, and some of them will simply leave.
Accessibility checks whether everyone can use your page, including people using screen readers or keyboard navigation. It looks at things like colour contrast, button labels, and image descriptions. Small fixes here help real people, not just the score.
Best Practices checks general technical health and safety, such as using HTTPS, avoiding outdated code, and showing images at the right size. Think of it as a basic hygiene check for your page.
SEO checks whether search engines can read the page properly. It looks for a title, a meta description, readable text, and links that can be crawled. A high SEO score here is a good start, though it does not guarantee rankings on its own.
The official PageSpeed Insights homepage at pagespeed.web.dev, where every speed test starts.
Lab Data vs Field Data, in Simple Words
Your report shows two kinds of numbers, and confusing them is very common. Field data is what real visitors experienced on your site over the last 28 days. It comes from real phones, real networks, and real locations. If your site gets enough traffic, this section decides whether you pass or fail the Core Web Vitals assessment.
Lab data is a single test run in controlled conditions, like a car tested on a test track instead of on real roads. It uses a simulated phone and a throttled connection, and it gives the same conditions every time. That makes it perfect for spotting problems and testing fixes, even on a brand new site with no visitors yet.
Use both together. Field data tells you what is actually happening to your visitors. Lab data tells you why, and lets you check a fix immediately instead of waiting 28 days to see if field data improves.
The Core Web Vitals: LCP, INP and CLS
Google groups the most important speed and stability measures under one name, the Core Web Vitals. The official web.dev guide to Web Vitals explains them in depth, but here is the simple version.
Largest Contentful Paint, or LCP, measures how long the biggest element on the screen takes to appear, usually the main image or the big heading. Good is 2.5 seconds or less. A slow LCP is often caused by a huge hero image or a slow server.
Interaction to Next Paint, or INP, measures how quickly the page responds when a visitor clicks, taps, or types. Good is 200 milliseconds or less. Heavy JavaScript running in the background is the usual troublemaker here.
Cumulative Layout Shift, or CLS, measures how much the page jumps around while loading. Good is 0.1 or less. You know this problem well: you go to tap a button, an advert loads above it, everything shifts, and you tap the wrong thing. Yes, Google scores your site for that frustration.
| Metric | Good | Needs Work | Poor |
|---|---|---|---|
| LCP (loading) | 2.5s or less | 2.5s to 4s | Over 4s |
| INP (responsiveness) | 200ms or less | 200ms to 500ms | Over 500ms |
| CLS (stability) | 0.1 or less | 0.1 to 0.25 | Over 0.25 |
Common Red Warnings and Plain Fixes
Huge images. The report may say "Properly size images" or suggest modern formats. Photos straight from a phone can be several megabytes. Resize images to the size they are actually shown, compress them, and use WebP where you can. This is usually the single biggest win.
Unused JavaScript. Warnings like "Reduce unused JavaScript" mean your page loads code it never uses, often from plugins or themes you half remember installing. Remove plugins you do not need, and delay scripts that are not needed for the first screen.
Slow server response. If the report flags a high server response time, your hosting is answering slowly. Cheap, overloaded hosting is a common cause. A good cache plugin helps, and sometimes the honest fix is better hosting.
Render-blocking resources. This means CSS or JavaScript files are pausing the page before anything can show. Loading non-essential files later, and keeping only critical styles up front, clears this warning in most cases.
If your site runs on WordPress, our step-by-step guide on how to speed up your WordPress website covers these fixes in practical order, from caching to image compression. Work through the PageSpeed list one warning at a time, re-test after each fix, and do not panic about reaching exactly 100. A solid green score in the 90s with fast real timings is a great result.
Frequently Asked Questions
Is Google PageSpeed Insights free to use?
Yes, it is completely free. You can test any public page as many times as you like, without signing in or paying anything.
Why is my mobile score lower than my desktop score?
The mobile test uses a slower simulated phone and a slower connection, which matches real visitors. That is why mobile scores are almost always lower, and why you should improve the mobile result first.
Why does my score change every time I run the test?
Small changes are normal. Network conditions and page elements like adverts or videos can vary slightly between runs. Test two or three times and judge by the typical result, not the best or worst one.
Do I need a score of 100 for good SEO?
No. Google uses the Core Web Vitals as one ranking signal among many. Passing the three vitals with comfortable margins matters far more than chasing a perfect 100 while ignoring your content.
Conclusion
PageSpeed Insights turns "my site feels slow" into clear numbers and a fix list. Run the test, check the mobile tab first, understand the four scores, watch LCP, INP, and CLS, and fix the red warnings one by one. Re-test after every change, and within an afternoon you will know exactly how fast your website really is.
Top comments (0)