Free website speed checker
Test your page with Google Lighthouse in a real Chrome browser and see how quickly it shows content, how long it stays unresponsive while loading, and which images, scripts or files are slowing it down.
What this tool checks
Lighthouse performance score
Lighthouse's own 0–100 score, a weighted combination of the loading metrics below. This is the score you'll see for this tool.
Largest Contentful Paint (LCP)
How long until the biggest piece of content — usually a hero image or headline — appears. Good is 2.5 seconds or less; over 4 seconds is poor.
First Contentful Paint (FCP)
How long until anything at all appears on the screen. Good is 1.8 seconds or less.
Total Blocking Time (TBT)
How long the page is too busy running scripts to respond to a tap or click during loading. Good is 200 ms or less.
Cumulative Layout Shift (CLS)
How much the content jumps around while the page loads. Good is 0.1 or less.
Speed Index and Time to Interactive
How quickly the visible page fills in (good: 3.4 s or less), and when it becomes reliably usable (good: 3.8 s or less).
What's slowing it down
Up to 15 failing Lighthouse audits, worst first, each naming up to five of the files or elements responsible with the kilobytes or milliseconds you could save. Passed audits are listed too.
Quick check when Lighthouse can't run
If Lighthouse is unavailable or returns no usable report, we fall back to a request-based quick check: server response, HTML size, compression, render-blocking files, caching headers and image size and format. It's labelled as such and isn't a Lighthouse score.
How the check works
We run Lighthouse's performance test once against your page in headless Chrome. By default Lighthouse simulates a mid-range phone on a slower mobile connection, so results are usually slower than what you see on a fast computer — that's deliberate, because many visitors aren't on fast devices. Each metric is compared with its published “good” and “poor” thresholds and marked pass, needs work or fail.
It's a lab test of a single page load, and every metric is labelled as such. Results vary between runs because network conditions, server load, ads and third-party scripts change from one load to the next, so a difference of a few points isn't meaningful. Compare several runs before drawing conclusions.
A lab test can't measure Interaction to Next Paint (INP), because INP needs real people tapping and clicking; Total Blocking Time is the closest lab signal. Real-visitor (field) data, like the Core Web Vitals report in Google Search Console, can differ from these lab results. When the quick check runs instead of Lighthouse, it makes one page request and at most 25 HEAD requests for your files and images.
Evidence you'll see
Each metric shows the measured value next to its threshold, labelled “Lab (Lighthouse, one run)”. Each slow audit names the files or elements behind it — for example the largest images or the render-blocking scripts — with the potential savings in KB or milliseconds. Findings come from Lighthouse, or from our own requests when the quick check ran.
What this tool does not check
Real-visitor (field) data
It's one lab run from our server. It doesn't include Chrome UX Report data or how fast the page is for your actual visitors.
Interaction to Next Paint (INP)
INP needs real interactions, so no lab test can measure it. Total Blocking Time is the closest stand-in.
Other pages and user journeys
Only the page you enter is loaded, once. Checkout flows, logged-in pages and repeat visits with a warm cache aren't measured.
A desktop run
Lighthouse's default mobile simulation is used; there's no separate desktop test.
Common problems we find
Oversized images
Photos uploaded straight from a camera or phone and shown much smaller than their real size.
Render-blocking CSS and JavaScript
Files in the page head that must finish downloading before anything can be shown.
Unused JavaScript and CSS
Large theme, plugin or framework bundles where most of the code isn't used on this page.
Heavy third-party scripts
Chat widgets, tag managers, video embeds and ad scripts that keep the browser busy.
Slow server response
The first byte arrives late because of slow hosting, no caching or heavy database work.
Layout shifts
Images and embeds without reserved space, and banners that push content down after it appears.
How to fix them
Resize and compress images
Export images at the size they're displayed (up to twice that for sharp screens), use WebP or AVIF, and lazy-load images below the first screen.
Defer scripts that aren't needed straight away
Add defer or async to non-essential scripts, remove plugins you don't use, and load chat or video widgets only when someone interacts with them.
Cache pages and use a CDN
Page caching and a content delivery network cut server response time, especially for visitors far from your server.
Reserve space for content
Give images and embeds width and height attributes or a CSS aspect-ratio so nothing jumps when they load.
Compress text files
Turn on gzip or Brotli compression for HTML, CSS and JavaScript in your server or hosting settings.
Frequently asked questions
Why is my score different from PageSpeed Insights?
PageSpeed Insights also uses Lighthouse, but runs it on different machines in a different location, and each run is a new page load. It also shows real-visitor data from the Chrome UX Report when there's enough traffic, which our lab test doesn't include.
Why does my score change every time I test?
Each test is a single page load, and network speed, server load and third-party scripts differ from one load to the next. Small swings are normal; look for consistent patterns across several runs.
Does this measure Core Web Vitals?
It measures LCP and CLS in the lab. The third Core Web Vital, INP, needs real interactions from real visitors, so it can't be measured in a lab test; Total Blocking Time is the closest stand-in. For field data, use the Core Web Vitals report in Google Search Console.
Is this a mobile or desktop test?
Lighthouse's default settings, which we use, simulate a mid-range phone on a throttled mobile connection. Pages that are fast on a desktop computer can still score lower here.
What happens if Lighthouse can't run?
We run a quick check instead: one request for the page plus a limited number of HEAD requests for its files and images. It reports server response, compression, render-blocking files, caching and image weight, and its score is clearly labelled as a quick check, not a Lighthouse score. If even the page request fails, we say so rather than showing a made-up score.
Helpful guides
- How to improve Core Web Vitals: a practical guide to LCP, INP and CLSThe three Core Web Vitals, the thresholds that count as good, why lab and field numbers differ, and the specific changes that improve loading, responsiveness and visual stability.Read the guide
- Why is my website slow? How to find the cause and fix it"Slow" can mean a slow server, a slow main image, a sluggish response to taps, or a page that jumps around. Here's how to tell which, and what to fix first.Read the guide
- How to reduce image size without losing visible qualityImages are often the heaviest part of a page. Resize, choose the right format, compress, and serve the right size to each device, step by step.Read the guide
Related free tools
- Mobile CheckerSee your page on phone, tablet and desktop screens and find content that spills off the screen.Open Mobile Checker
- Website AuditCheck one page for SEO, speed, mobile, accessibility, security, broken links and server problems in a single report.Open Website Audit
- SEO CheckerCheck your page title, description, headings, image descriptions, canonical URL, robots.txt and sitemap.Open SEO Checker
Rather have someone fix it for you?
These checks are free and so are the guides. If you'd prefer a person to make the changes, our team offers paid help.