Skip to content

How to test Core Web Vitals: free tools and how to read the results

A practical walkthrough of every free way to test Core Web Vitals, why field and lab results disagree, and what a pass or fail really means for your pages.

The seoscore team 10 min read

On this page
  1. What a Core Web Vitals test checks
  2. Field data vs lab data: read this before you test
  3. How to run a Core Web Vitals test in PageSpeed Insights
  4. How to check Core Web Vitals for the whole site in Search Console
  5. How to test Core Web Vitals in Chrome DevTools and Lighthouse
  6. Measuring Core Web Vitals from real visitors yourself
  7. Which Core Web Vitals test to use
  8. Why your Core Web Vitals results don't match between tools
  9. Your test failed: what to fix first
  10. Retest after you fix
  11. Check your whole site with seoscore
  12. Next steps

To test Core Web Vitals, enter a URL in PageSpeed Insights and read the "Core Web Vitals Assessment" box at the top. That verdict comes from real Chrome users over the last 28 days, which is the data Google uses. A page passes only when LCP, INP and CLS are all in the "good" range at the 75th percentile of visits. For the whole site, use the Core Web Vitals report in Google Search Console. To find out why a page is slow, use Chrome DevTools or Lighthouse.

What a Core Web Vitals test checks

Core Web Vitals are three measurements of how a page feels to use (web.dev):

  • Largest Contentful Paint (LCP): how long the biggest piece of content in view, usually a hero image or headline, takes to appear. It measures loading.
  • Interaction to Next Paint (INP): how quickly the page visibly responds after a click, tap or key press, across the whole visit. It measures responsiveness.
  • Cumulative Layout Shift (CLS): how much the content jumps around while the page loads. It measures visual stability.

Each metric has three bands (web.dev):

Metric Good Needs improvement Poor
LCP 2.5 s or less 2.5 s to 4 s Over 4 s
INP 200 ms or less 200 ms to 500 ms Over 500 ms
CLS 0.1 or less 0.1 to 0.25 Over 0.25

The value that counts is the 75th percentile: the figure that three in four visits are at or under. If 75% of visits see the main image within 2.2 seconds, LCP is good, even if a few visitors on slow phones waited six. Google picked the 75th percentile because it is less affected by outliers. A page or site passes the assessment only when the 75th percentiles of all three metrics are good (PageSpeed Insights documentation).

INP replaced First Input Delay (FID) as a Core Web Vital in March 2024. If a guide or tool still reports FID, it is out of date.

On rankings, Google's page experience guidance says Core Web Vitals are used by its ranking systems, but that a good page experience doesn't make up for content that isn't relevant or helpful. Passing is worth getting right; it won't carry a weak page on its own.

Field data vs lab data: read this before you test

Most confusion about Core Web Vitals comes from mixing up two kinds of data.

Field data is measured on real visits. Google collects it from Chrome users into the Chrome UX Report (CrUX), anonymises it and reports it over a rolling 28-day window. It covers many devices and network conditions. This is what the PageSpeed Insights assessment and Search Console show, and what Google uses.

Lab data is a single simulated page load on a fixed device and network, usually run by Lighthouse. It is instant and repeatable, which makes it ideal for debugging and for checking a fix. But it is one load on one setup, not your audience.

INP shows the gap most clearly. INP needs a person clicking, tapping and typing. Lighthouse loads the page without touching it, so it can't measure INP. Instead it reports Total Blocking Time (TBT): how long the browser's main thread was blocked by long tasks during loading. A low TBT usually helps INP, but they are different measurements, and a page can have a good TBT and a poor INP if, for example, a menu or filter runs heavy JavaScript when clicked.

How to run a Core Web Vitals test in PageSpeed Insights

PageSpeed Insights (PSI) is the quickest free Core Web Vitals check for one page, because it shows field and lab data together.

1. Enter the URL

Go to PageSpeed Insights, paste the full URL of the page, including https://, and run the test.

2. Read the Core Web Vitals Assessment

The box at the top says Passed or Failed, then lists each metric with its 75th percentile value and a bar showing the share of visits that were good, needed improvement or were poor. This is field data.

Two rules from the PSI documentation explain results that look odd. If there isn't enough data for INP, the page can still pass on LCP and CLS alone. If LCP or CLS lacks data, the page can't be assessed at all.

3. Switch between this URL and origin, mobile and desktop

Tabs let you switch between mobile and desktop, which are assessed separately. A toggle lets you switch between data for this URL and for the origin, meaning the whole domain. When a URL doesn't have enough data of its own, PSI falls back to origin data where it exists. That tells you about the site on average, not about this page.

4. Read the lab section below

Further down is the Lighthouse lab test: a performance score out of 100, lab metrics including LCP, CLS and TBT, and a list of suggested improvements. The score out of 100 is not your Core Web Vitals result. A page can score 95 in the lab and fail the field assessment, or the reverse.

When PSI says there is no data

Low-traffic pages often get a message that the Chrome UX Report doesn't have enough real-world data. Sometimes the origin has no data either, and PSI shows no field data at all. Google publishes no minimum traffic figure. If you see this:

  • check the origin view for a site-wide picture
  • check Search Console, which groups similar pages and may have data for the group
  • use lab tests to find problems
  • collect your own field data (see below)

How to check Core Web Vitals for the whole site in Search Console

PSI checks one URL. The Core Web Vitals report in Google Search Console shows the whole site, using the same field data.

  1. Open your property and choose Core Web Vitals under Experience in the left-hand menu.
  2. You'll see separate charts for mobile and desktop, with counts of URLs rated Poor, Need improvement and Good. Open the report for one device.
  3. The issues table lists each problem by metric and threshold missed, such as LCP above 2.5 seconds on mobile.
  4. Click an issue to see the affected URL groups. Search Console groups similar pages, such as all product pages built from one template. A group's status is set by its worst metric, and each value is a 75th percentile over the last 28 days.
  5. Open a group to see example URLs, then run one through PSI and DevTools to find the cause.
  6. After you fix it, click Validate fix. Google then keeps evaluating the URLs for 28 days, with statuses of Pending, Passed or Failed. Pending means it is waiting for enough data.

If a group lacks data, Search Console rolls it into a broader origin group. If the site as a whole doesn't get enough Chrome traffic, the report shows a warning that there isn't enough usage data for that device type and stays empty. That says nothing good or bad about your speed; there just isn't enough data.

How to test Core Web Vitals in Chrome DevTools and Lighthouse

Live metrics in the Performance panel

Open the page in Chrome, open DevTools (F12, or Cmd+Option+I on a Mac) and choose the Performance panel. The Live Metrics screen shows LCP, CLS and INP for your own visit as it happens (Chrome DevTools documentation).

  • Reload the page to see LCP and CLS.
  • Click buttons, open menus and type into forms to see INP. Each interaction that counts towards INP appears in an interaction log, and interactions over 200 ms are flagged.
  • Switch on field data to see your site's CrUX figures next to your local ones, for this URL or the origin, on mobile or desktop.

This is the best free way to find which interaction makes INP poor. Google now recommends the Performance panel instead of the older Web Vitals browser extension.

Run Lighthouse

In DevTools, open the Lighthouse panel, pick Mobile or Desktop, tick Performance and analyse the page load. You get the same kind of lab report PSI shows, run from your own machine.

Settings that change your results

  • Throttling: Lighthouse's mobile test simulates a slower phone and network. Live Metrics uses your real computer unless you add CPU and network throttling in the Performance panel, so a fast laptop flatters the numbers.
  • Device: mobile and desktop emulation give different results. Test mobile first.
  • Extensions: ad blockers and other extensions change what loads. Test in an Incognito window, where extensions are usually off.
  • Cache: a repeat visit with a warm cache is faster than a first visit.
  • Variance: run a lab test three times and use the middle result.

Measuring Core Web Vitals from real visitors yourself

If your pages don't have CrUX data, or you want faster feedback than a 28-day window, measure real visits yourself with Google's open-source web-vitals JavaScript library:

import {onCLS, onINP, onLCP} from 'web-vitals';

function sendToAnalytics(metric) {
  const body = JSON.stringify({name: metric.name, value: metric.value, id: metric.id});
  (navigator.sendBeacon && navigator.sendBeacon('/analytics', body)) ||
    fetch('/analytics', {body, method: 'POST', keepalive: true});
}

onCLS(sendToAnalytics);
onINP(sendToAnalytics);
onLCP(sendToAnalytics);

Send the values to your analytics tool, then report the 75th percentile per page template and per device, so your numbers line up with Google's.

For history, Google's CrUX tools, including CrUX Vis and the CrUX API, show how an origin's or URL's field metrics have moved over time. That helps you see whether a dip started with a particular release.

Real-user monitoring is worth the effort when you have important low-traffic pages, when INP is failing and you need to know which interactions cause it, or when you ship often and can't wait four weeks to learn whether a change helped.

Which Core Web Vitals test to use

Tool Data Scope INP Best used for
PageSpeed Insights Field and lab One URL, plus origin field data Field only (TBT in lab) A quick pass or fail for one page
Search Console Field Whole site, grouped by similar pages Yes Finding which templates fail; validating fixes
DevTools Performance panel Your own visit, plus CrUX comparison One page Yes, as you interact Debugging the cause, especially INP
Lighthouse Lab One page No (uses TBT) Repeatable before and after checks
CrUX tools Field Origin or URL Yes Trends over time
web-vitals library Your own field data Every page you add it to Yes Pages without CrUX data; faster feedback

Why your Core Web Vitals results don't match between tools

  • Field vs lab: real visitors on mid-range phones versus one simulated load.
  • The 28-day window: field data includes visits from before your latest change.
  • Mobile vs desktop: they are measured and reported separately.
  • URL vs origin: PSI may be showing the whole domain because the page lacks data.
  • Groups: Search Console reports a group's worst metric across many pages, while PSI reports one URL.
  • Location and device: your test runs from one place on one machine; your visitors don't.
  • INP vs TBT: the lab can't measure INP, so it shows a different metric.

Which to trust? For "do I pass?", trust field data: Search Console for the site, the PSI assessment for a page. For "why?" and "did my change work?", use lab tests.

Your test failed: what to fix first

Start with the worst metric (poor before needs improvement) on your most-visited page templates, such as the homepage, product pages or blog posts. Fixing a template fixes every page built from it.

LCP

  • Slow server response: add page caching, use a CDN, or upgrade slow hosting.
  • Render-blocking resources: CSS and JavaScript in the page head that must load before anything shows. Defer scripts that aren't needed straight away.
  • The hero image: compress it, serve a modern format at the size it is displayed, don't lazy-load it, and give it priority:
<img src="hero.webp" width="1200" height="600" fetchpriority="high" alt="Our workshop">

INP

  • Long JavaScript tasks: work that blocks the main thread for more than 50 ms. Break it up or postpone what isn't needed for the next frame.
  • Third-party scripts: chat widgets, tag managers, ad and tracking scripts. Remove the ones you don't use and load the rest later.
  • Heavy click handlers: do the visible update first and the rest afterwards.

CLS

  • Images, videos and iframes without dimensions: add width and height attributes so the browser reserves space.
  • Ads, embeds and banners: reserve their space with a fixed minimum height, and don't insert content above what is already on screen.
  • Web fonts: a late-loading font can reflow text. Preload key fonts and choose a fallback font of similar size.

Core Web Vitals are one part of technical SEO; our guide to technical SEO covers the rest.

Retest after you fix

Confirm the change straight away with a lab test: run Lighthouse or watch Live Metrics before and after. Field data takes longer. Because CrUX uses a rolling 28-day window, the full effect can take up to four weeks to show in PSI's assessment and in Search Console, and Validate fix monitors URLs for 28 days.

Then keep checking. A new plugin, a theme update, a tag added for a campaign or a chat widget can undo the work without anyone noticing.

Check your whole site with seoscore

A Core Web Vitals test checks one URL at a time, and speed is only part of site health. The seoscore SEO audit checks your site's pages and scores the site out of 100 across eight weighted areas. Performance is 16 of those points, alongside Content Quality, Technical, Mobile, Security, Accessibility, Social and AEO. Each fix is ranked by the points it adds, so you can see whether a speed problem is worth more than, say, missing page titles, and fix in that order. Every check and weight is published on the methodology page.

To catch regressions, automatic audits re-audit the site weekly or daily and email the team when the score drops or the site can't be reached. They are on Starter and up, with a free trial. On WordPress, the WordPress plugin shows the score and fixes inside wp-admin and audits each page as it is edited.

Use seoscore to decide what to fix first and to spot drops, and PSI and Search Console to confirm your Core Web Vitals status itself.

Next steps

  1. Run your homepage and one page from each main template through PageSpeed Insights on mobile. Note pass or fail and the worst metric.
  2. Open the Core Web Vitals report in Search Console to see which page groups fail.
  3. Reproduce the problem in the DevTools Performance panel and fix the worst metric on your busiest template.
  4. Confirm with a lab test, click Validate fix, and check the field data again after 28 days.
  5. Set up regular checks so the next plugin or script doesn't quietly undo the work.

Questions people ask

Does a Lighthouse score of 100 mean I pass Core Web Vitals?

No. The Lighthouse score is a lab result from one simulated load and combines several metrics, not just the Core Web Vitals. Google assesses Core Web Vitals on field data from real visitors, so a page can score highly in Lighthouse and still fail, or score modestly and pass.

Why does my Lighthouse score change every time I run it?

Server response times, network conditions, background apps on your computer, browser extensions and third-party scripts all vary between runs. Run the test three times in an Incognito window and use the middle result. Small swings of a few points are normal and not worth chasing.

Do I need to pass Core Web Vitals on both mobile and desktop?

They are measured and reported separately, so a page can pass on one and fail on the other. Most sites get more of their traffic on phones, and phones are slower, so mobile is usually the one to fix first. Search Console shows both so you can track each.

Is a free Core Web Vitals test enough, or do I need a paid tool?

For most sites the free tools are enough: PageSpeed Insights and Search Console give you Google's own field data, and DevTools and Lighthouse show the causes. Paid tools mainly add convenience, such as monitoring many pages on a schedule, alerts and history. Real-user monitoring helps most on busy sites that ship changes often.

Sources

  1. How the Core Web Vitals metrics thresholds were defined  |  Articles  |  web.dev web.dev
  2. Web Vitals  |  Articles  |  web.dev web.dev
  3. About PageSpeed Insights  |  Google for Developers developers.google.com
  4. developers.google.com/search/docs/appearance/page-experience developers.google.com
  5. Core Web Vitals report - Search Console Help support.google.com
  6. Performance features reference  |  Chrome DevTools  |  Chrome for Developers developer.chrome.com
  7. developer.chrome.com/docs/crux developer.chrome.com

seoscore

What does your site score?

Run a free audit: a score out of 100 and every fix ranked by the points it adds.

Get my free score
More on Site speed

Technical SEO

What is technical SEO? Elements, examples and a checklist

Technical SEO is the work that lets search engines and AI crawlers find, render, understand and index your pages. Here is what it covers, how to spot the common problems and what to fix first.

12 min read