PageSpeed & Core Web Vitals Test

Enter a page address. The test runs Google's PageSpeed Insights engine for mobile and desktop at the same time and shows the scores, the Core Web Vitals, the element that decides your LCP, and what to fix first.

FreeNo sign-upRuns in your browser: your data is not uploaded

Built by Faizan · Updated 22 Oct 2026

What the test measures

The report is produced by Google's PageSpeed Insights API, the same Lighthouse engine behind pagespeed.web.dev, run twice: once as a mid-range Android phone on a slow 4G connection (mobile) and once as a desktop browser. Two things come back:

Lab data: a fresh load of the page under controlled conditions. The Performance score (0-100) is computed from five metrics: First Contentful Paint, Largest Contentful Paint, Total Blocking Time, Cumulative Layout Shift and Speed Index. Accessibility, Best practices and SEO scores come from the same run.
Field data (Core Web Vitals): what real Chrome users experienced on that page over the last 28 days, at the 75th percentile. It appears only for pages with enough traffic; a new site shows "no field data" until Google has collected it. This is the data Google's ranking systems use.

Core Web Vitals thresholds

Metric	Good	Needs improvement	Poor	What it means
LCP	≤ 2.5 s	2.5-4.0 s	> 4.0 s	When the largest image or text block above the fold is painted
INP	≤ 200 ms	200-500 ms	> 500 ms	How long the page takes to react to taps and clicks (field only)
CLS	≤ 0.1	0.1-0.25	> 0.25	How much the layout jumps while loading
TBT	≤ 200 ms	200-600 ms	> 600 ms	Lab stand-in for INP: time the main thread was blocked by scripts

How to read the fixes

The "What to fix" list is sorted by the time and bytes each item would save, so start at the top. Each entry names the audit Lighthouse ran, the files or elements involved, and in plain words what usually causes it on a PHP, WordPress or Laravel site. The LCP element line shows the exact HTML element that decided your Largest Contentful Paint; nine times out of ten the fix is to get that image into the HTML early (not injected by JavaScript, not loading="lazy"), preload it, and serve it at the size it is displayed.

Scores vary by a few points between runs; Google says so on every report. Judge a change by the metrics, and run the test two or three times before deciding. If the page is generated on every request (PHP, WordPress without a cache), the first run after a change can hit a cold server and read low; run it again.

Before you optimise

Test the pages that earn traffic, not only the home page: a blog post or product page often has different problems.
Fix in this order: LCP (image discoverability, render-blocking CSS, server response), then CLS (image dimensions, fonts, injected banners), then TBT/INP (third-party scripts, long tasks).
Third-party scripts (analytics, ads, chat widgets) are usually the bulk of unused JavaScript; load them after the page is interactive.
After fixing the lab metrics, wait a few weeks for the field data to follow; the field numbers are what count for ranking.

FAQ

Is this the same as Google PageSpeed Insights?

Yes. The test calls Google's PageSpeed Insights API, which runs Lighthouse on Google's servers. The scores and metrics are the ones you would see on pagespeed.web.dev; this page adds the mobile and desktop runs side by side, the LCP element, and a fix list sorted by saving.

Why is my mobile score so much lower than desktop?

Mobile is simulated on a mid-range phone over a slow 4G connection with a slower CPU, so every byte and every script costs several times more. A page that scores 95 on desktop and 60 on mobile usually has large images, render-blocking CSS or heavy JavaScript that a fast desktop hides.

Why do I get a different score each time?

Lighthouse simulates network and CPU throttling and the result varies by a few points; Google notes "values are estimated and may vary" on every report. Cached results from the previous six hours are shown for the same address so the number does not jump while you read it.

What does "no field data" mean?

Google only reports real-user Core Web Vitals for pages with enough Chrome traffic in the last 28 days. New or small sites have none; the lab metrics are the guide until field data appears in Search Console.

Does the PageSpeed score affect Google rankings?

The field Core Web Vitals (LCP, INP, CLS from real users) are a ranking signal within the page experience system. The lab score itself is a diagnostic; improving what it reports improves the field numbers over time.

Can I test a page that needs a login?

No. Google fetches the page as an anonymous visitor, so login pages, staging sites behind basic auth, and pages that block bots fail with "could not load that page".

Link to this tool

Found it useful? Put a link in your docs, README or forum answer so others can use it too:

HTML

<a href="https://headsntail.com/tools/page-speed-test">PageSpeed & Core Web Vitals Test</a> (free, by HeadsNTail)

Markdown (GitHub, Stack Overflow, dev.to)

[PageSpeed & Core Web Vitals Test](https://headsntail.com/tools/page-speed-test)

Need a website, HR or business system? HeadsNTail builds websites, HR and payroll apps and business tools. See services or ask a question.

More free tools

SEO Checklist (39 points, printable)Tick off technical SEO, schema, E-E-A-T content, internal links, indexing, AI search and measurement. Progress saves in your browser; print it as a PDF. Free SEO Audit ToolScore any website 0-100: HTTPS, sitemap, titles, thin content, E-E-A-T, schema, broken links, orphans, speed, security - with the fix for each. Days Between Dates & Working DaysTotal days, working days and hours between two dates with your own weekend (Fri-Sat, Fri only, Sat-Sun) and holidays. Time CalculatorAdd or subtract hours and minutes, hours between two times, and a timesheet total in decimal hours.