Guide
Google PageSpeed Insights: what your score means and how to fix it (2026)
The short answer
Google PageSpeed Insights is free and shows two things: a 0-100 lab score from one simulated load, and Core Web Vitals from real Chrome users over 28 days. The real-user numbers are what Google's ranking systems use, with good meaning LCP 2.5 s or less, INP 200 ms or less and CLS 0.1 or less. Good scores help but don't guarantee rankings; relevant content still wins.
What PageSpeed Insights tells you
Google PageSpeed Insights is a free tool that tests how fast a page loads and how stable it feels. Paste in a web address and it returns two kinds of results, per Google’s PageSpeed Insights documentation:
- Field data: how real Chrome users experienced the page over the previous 28 days, from the Chrome User Experience Report (CrUX). This is where your Core Web Vitals come from.
- Lab data: one simulated page load run by Lighthouse, Google’s open-source testing tool. This is where the 0-100 score comes from, along with scores for Accessibility, Best Practices and SEO.
Field data is what your visitors lived through. Lab data is a controlled test that helps you find the cause. A small or new site may show only lab data, because there isn’t enough real traffic yet for a field report.
What the score means
The big number at the top is the Lighthouse performance score. Google sorts it into three bands:
| Score | Rating | Color |
|---|---|---|
| 90-100 | Good | Green |
| 50-89 | Needs improvement | Orange |
| 0-49 | Poor | Red |
The score is a weighted mix of five lab measurements. In Lighthouse 10, Total Blocking Time counts for 30%, Largest Contentful Paint 25%, Cumulative Layout Shift 25%, First Contentful Paint 10% and Speed Index 10%, according to Chrome’s Lighthouse scoring page.
Why your score jumps around
Run the same page twice and you may get two different scores. Google says scores vary between runs because of things like A/B tests, ads, browser extensions, antivirus software and the test device. Test three times and look at the pattern, not one number.
Why mobile scores lower than desktop
PageSpeed Insights reports mobile and desktop separately. The mobile lab test emulates a mid-tier Moto G4 phone on a mobile network; the desktop test uses an emulated desktop on a wired connection. Lighthouse has used separate scoring curves for desktop since version 6, so a mobile score and a desktop score can’t be compared with each other.
Core Web Vitals: the numbers Google uses
Core Web Vitals are three measurements of how a page feels to a real visitor. These are the thresholds Google publishes:
| Metric | What it measures | Good | Poor |
|---|---|---|---|
| Largest Contentful Paint (LCP) | When the main content appears | 2.5 s or less | Over 4 s |
| Interaction to Next Paint (INP) | How fast the page reacts to taps and clicks | 200 ms or less | Over 500 ms |
| Cumulative Layout Shift (CLS) | How much the layout jumps while loading | 0.1 or less | Over 0.25 |
A page passes when the 75th percentile of visits is good on all three. If INP doesn’t have enough data, LCP and CLS decide it. INP replaced First Input Delay (FID) as a Core Web Vital on March 12, 2024, so guides that still mention FID are out of date.
Does the score affect your Google ranking?
Partly, and less than most people think. Google says Core Web Vitals are used by its ranking systems, per its page experience guide. The same page also says:
- There is no single page experience signal.
- Good Core Web Vitals results in Search Console or third-party tools don’t guarantee top rankings.
- Google always tries to show the most relevant content, even if its page experience is sub-par.
- Page experience helps most when many pages are similarly helpful.
Google’s Search Central page on Core Web Vitals still “highly recommends” good scores “for success with Search.” Our reading: speed is a tiebreaker. A slow page with the best answer can still outrank a fast page with a weak one. A fast page with a good answer is the goal.
Also note what Google’s ranking guidance talks about: Core Web Vitals, measured from real users. It doesn’t ask you to hit a Lighthouse score. Chasing 100 on mobile can waste time that would be better spent on content. Our guide to websites and SEO covers the parts that matter more.
Speed matters for ads too, because a slow landing page wastes clicks you paid for. See websites and Google Ads.
How to run a test
- Open PageSpeed Insights and paste the full address of the page.
- Click Analyze and wait for the results.
- Read the field data first, if it’s there. That’s the pass or fail that counts.
- Switch between the Mobile and Desktop tabs.
- Scroll to the diagnostics to see what slows the page down.
Test your home page, your most visited service or product page, and any page you send ads to. You can also run Lighthouse in Chrome: open DevTools on the page, click the Lighthouse tab, keep the categories enabled and run the audit. Google says it takes 30 to 60 seconds, per the Lighthouse overview.
For a site-wide view, Search Console has a Core Web Vitals report. Our Search Console guide shows how to set it up.
The fixes Google recommends
Google’s list of the most effective Core Web Vitals fixes comes down to a few moves for each metric:
- LCP (slow main content): make the main image or text discoverable from the HTML and load it first, aim for instant navigations, and use a CDN to cut server response time.
- INP (slow reactions): break up long tasks, avoid unnecessary JavaScript, and avoid large rendering updates.
- CLS (jumping layout): set explicit sizes on images, embeds and ads, make pages eligible for the browser’s back/forward cache, and avoid animations that move layout.
Fixes per builder
How much of that list you can act on depends on your builder. On hosted builders, the platform controls the code and the servers. You control what you put on the page.
Wix, Squarespace and Webflow
- Shrink images before you upload. A huge photo at the top of the page is a common cause of a slow LCP.
- Go easy on the hero video. A full-screen video background is often the biggest file on the page.
- Count your embeds. Chat widgets, social feeds, review widgets and booking widgets each add JavaScript, which Google lists as a cause of slow INP.
- Remove apps and code you no longer use. On Wix, each App Market app you add can mean more to load.
- Keep the top of the page simple. Pop-ups and banners that push content down hurt CLS.
Shopify
Apps are the usual suspect. Many Shopify apps run on monthly subscriptions, and each one you add can put more code on your storefront. Uninstall apps you don’t use, and check that their code is gone from your theme. Then work through images the same way as on any builder.
WordPress
WordPress gives you the most control, and the most ways to slow a site down. We build on WordPress every week. What makes the biggest difference on our sites is a lightweight theme, as few plugins as the job allows, a caching setup and a CDN in front of the site. Page builder plugins and sliders tend to add weight.
Your host matters more on WordPress than on any builder, because server response time feeds LCP. If a test flags slow server response, compare hosts in our web hosting guide. Site Kit by Google, the free official plugin, can show PageSpeed Insights results inside your WordPress dashboard.
Common mistakes
- Testing only the home page. Your service pages, product pages and ad landing pages are often heavier, and they’re the ones that earn money.
- Treating one lab run as the truth. Scores move between runs. Look at field data and at the trend over several tests.
- Comparing mobile with desktop. They use different scoring curves, so a gap between them is normal.
- Following old advice about FID. INP replaced it in March 2024.
- Adding a speed app to fix a speed problem. On hosted builders, removing things usually helps more than adding one more tool.
- Ignoring layout shift. A button that moves just as someone taps it is a CLS problem, and it costs you clicks even when the page loads fast.
What it costs, and when speed is a reason to leave
PageSpeed Insights and Lighthouse are free. Most fixes cost time, not money: smaller images, fewer apps, fewer widgets. On WordPress, the cheapest setup on this site is Bluehost Starter at $3.99 a month for 36 months ($143.64, renewing at $9.99), or $432.60 over 3 years with a domain ($36.96) and one Google Workspace inbox ($252).
If your builder’s own code is the bottleneck and you’ve already trimmed what you control, moving is the remaining option. Check our exit scores first: self-hosted WordPress scores 10/10, Squarespace and Shopify 5/10, Wix 3/10. Then read leaving your website builder before you commit.
What to do next
- Back to the full list of free website tools.
- Track real visitors with Google Analytics.
Questions people ask
What is a good PageSpeed Insights score?
Google rates 90-100 as good, 50-89 as needs improvement and 0-49 as poor. The Core Web Vitals pass or fail from real users matters more than the score.
Why is my mobile score so much lower than desktop?
The mobile test emulates a mid-tier Moto G4 on a mobile network, and Lighthouse uses separate scoring curves for desktop and mobile. The two scores can't be compared.
Does PageSpeed affect SEO?
Google says Core Web Vitals are used by its ranking systems, but good results don't guarantee top rankings and the most relevant content still wins. Speed helps most when competing pages are similarly helpful.
Why does my score change every time I test?
Google says results vary because of A/B tests, ads, browser extensions, antivirus software and the test device. Run a few tests and look at the pattern.
Why does PageSpeed Insights show no field data for my site?
Field data comes from real Chrome users over the previous 28 days. Small or new sites often don't have enough visits yet, so only the lab test appears.
Keep reading
Would you rather have someone build it?
We also build and look after WordPress sites for US businesses. Flat fees from $700. This is our own service. It isn’t part of our rankings.