Three metrics, one sentence each
| Metric | What it measures | Good | Needs improvement | Poor |
|---|---|---|---|---|
| LCP (Largest Contentful Paint) | When the largest element on screen appears, usually an image or a heading | up to 2.5 s | 2.5 to 4 s | over 4 s |
| INP (Interaction to Next Paint) | How fast the page responds to clicks and taps | up to 200 ms | 200 to 500 ms | over 500 ms |
| CLS (Cumulative Layout Shift) | How much elements move around while loading | up to 0.1 | 0.1 to 0.25 | over 0.25 |
Google judges a page by the 75th percentile, meaning the result should be good for at least three quarters of visits. Mobile counts most, because that's where pages are slowest.
How to check your site in 5 minutes
- Open PageSpeed Insights
Enter the page address at pagespeed.web.dev.
- Look at the top part first
The real-user data comes from actual visitors over the last 28 days. It appears when the site has enough traffic.
- Then at the diagnostics
The lab test shows specific causes: oversized images, render-blocking scripts, layout shifts.
- Check a few pages
The home page is often fast, while a services page with a gallery isn't.
The most common causes and fixes
| Cause | Metric | Fix |
|---|---|---|
| A large image at the top of the page | LCP | WebP or AVIF format, the right size for phones, no lazy loading for that one image |
| Fonts from an external server | LCP, CLS | Fonts on your own server, only the weights you need, font-display: swap |
| Third-party scripts: chat, pixels, widgets | INP, LCP | Remove the unused ones, load the rest later or after interaction |
| Images and frames without dimensions | CLS | Give the width and height, or the aspect ratio |
| A consent banner pushed in above the content | CLS | The banner as an overlay instead of pushing content down |
| A heavy theme and plugins | All | A lighter theme, fewer plugins, server-side caching |
Start with the list of third-party scripts. Every widget added at some point and no longer used slows the page down for every visitor.
Why it matters beyond Google
Speed isn't just about your position in search. A slow site on a phone loses visitors before they see the offer, and a jumping layout causes accidental clicks. It's usually the cheapest thing to fix that genuinely affects the number of enquiries.
Terms from this guide
- Core Web Vitals
Three Google metrics describing a page's speed and stability: LCP, INP and CLS.
- LCP
Largest Contentful Paint: the time after which the largest element visible on screen appears. Good when it doesn't exceed 2.5 seconds.
- INP
Interaction to Next Paint: how fast a page responds to clicks and taps. Good when it doesn't exceed 200 milliseconds.
- CLS
Cumulative Layout Shift: a measure of how much page elements move while loading. Good when it doesn't exceed 0.1.
Questions and answers
Is a score of 100 in PageSpeed necessary?
No. The lab score is a hint. What matters more is data from real users and whether the metrics are within the thresholds.
Why does the score change with every test?
The lab test depends on the momentary load on the network and the server. Compare several measurements or the 28-day user data.
Does site speed affect positions in Google?
Yes, Core Web Vitals are one of the signals used to assess a page, although the content and how well it matches the query are more important. Speed also strongly affects whether a visitor stays.
Sources and documentation
- web.dev: Core Web Vitalsweb.dev
- web.dev: LCPweb.dev
- web.dev: INPweb.dev
- web.dev: CLSweb.dev
- PageSpeed Insightspagespeed.web.dev
- Search Console: the Core Web Vitals reportsupport.google.com