A slow site loses visitors before they read a word, and speed is one of the signals Google uses for page experience. The good news is that the best tools for measuring it are free, and most fixes fall into a few familiar groups. This guide covers which tools to use, what the numbers mean and where to start.
Lab data and field data
Speed tools give you two kinds of numbers. Lab data comes from a test run on a simulated phone or computer. It is repeatable and shows the result of a change at once. Field data comes from real visitors using Chrome, collected in the Chrome User Experience Report. It shows what people actually experience, but it covers the last 28 days, so changes take weeks to show up. Field data is what Google uses. Lab data is how you find and test fixes.
Core Web Vitals
Google measures page experience with three Core Web Vitals:
- Largest Contentful Paint (LCP): how long the biggest image or block of text takes to appear.
- Interaction to Next Paint (INP): how quickly the page responds when someone taps, clicks or types. INP replaced First Input Delay (FID) in March 2024, and it is stricter because it looks at every interaction during a visit.
- Cumulative Layout Shift (CLS): how much the page jumps around while it loads.
The free tools worth using
All of these are free to use for testing individual pages.
| Tool | What it shows | Best for |
|---|---|---|
| PageSpeed Insightspagespeed.web.dev | Real-visitor data from Chrome when available, plus a Lighthouse lab test | The first check of any page |
| Search ConsoleCore Web Vitals report | Groups of your pages that pass or fail, from real visits | Seeing the whole site at once |
| LighthouseIn Chrome DevTools | A lab test with a list of fixes | Testing changes before they go live |
| Chrome DevToolsPerformance panel | Live LCP, INP and CLS as you click around, and what caused them | Finding slow interactions |
| WebPageTestwebpagetest.org | Filmstrips, request waterfalls, many locations and devices | Deep dives into how a page loads |
| CrUX Viscruxvis.withgoogle.com | Weekly real-visitor trends for an origin or page | Checking whether a fix helped over time |
Google retired its older CrUX Dashboard after November 2025 and recommends CrUX Vis in its place.
How to test properly
- Test on mobile first. Most tools default to a mid-range phone on a slower connection, which is closer to many real visitors than your office computer.
- Run each lab test three times and look at the middle result. Scores move between runs.
- Test the pages that matter, such as a product page, a blog post and the contact page, as well as the home page.
- Treat the Lighthouse score as a guide. The Core Web Vitals and the list of opportunities matter more than the number at the top.
The fixes that matter most
Images
- Resize images to the size they are shown at, and serve modern formats such as WebP or AVIF.
- Set
widthandheighton every image so the browser can reserve space. This fixes a lot of CLS. - Lazy-load images further down the page with
loading="lazy", but never the main image at the top. Give that onefetchpriority="high"to speed up LCP.
Server and caching
- A slow server response delays everything. Page caching, a current PHP version and a host near your visitors all help.
- A content delivery network serves files from locations close to each visitor.
- Set long cache lifetimes for images, styles and scripts so repeat visits load fast.
Scripts and plugins
- Remove plugins and scripts you no longer use. Each one adds code to every page.
- Load scripts with
deferwhere possible so they do not block the page. - Review third-party tags such as chat widgets, ad scripts and trackers. They are a common cause of poor INP.
- Break up long-running JavaScript so the browser can respond to taps between tasks.
Fonts and layout
- Host fonts yourself, load only the weights you use and set
font-display: swap. - Reserve space for ads, embeds and banners so they do not push content down when they load.
Common questions
- What score should I aim for?
- Aim for all three Core Web Vitals in the Good band in field data. A Lighthouse score in the 90s is a nice sign, but passing Core Web Vitals for real visitors matters more.
- Why does PageSpeed Insights say there is no field data?
- Chrome only reports data for pages and sites with enough visits. Until then, rely on the lab results.
- Why do my results change each time?
- Lab tests depend on network conditions and server load at that moment. Run several tests and compare the middle result.
- Will a faster site rank higher?
- Page experience is one of many signals Google uses, and relevant content matters more. A fast site does keep more visitors, whatever it does for rankings.
Speed work is a series of small, measurable fixes. If you would like someone to run the tests and make the changes, a professional on Internetivo can take it on as a project. Here is how your first project works.