Google Website Speed Test: How to Check and Improve Your Site Speed

A Google website speed test measures how quickly your website loads, responds to user interactions, and stays visually stable. You can test these using four free Google tools: PageSpeed Insights, Lighthouse, Chrome User Experience Report (CrUX), and Google Search Console. Together, they reveal performance problems and how real visitors experience your website.

Website speed directly affects user experience, conversion rates, and search performance, which makes measuring website speed, performance and KPIs hugely important. But there's a catch: a website that scores 100 in a synthetic test can still frustrate real customers.

That's why Google offers different ways to measure website performance. Some help developers diagnose problems. Others show whether those problems affect real users.

Google PageSpeed Insights is the easiest way to check website speed without installing anything.

Here's how to use it:

1. Open PageSpeed Insights.
2. Enter your website URL.
3. Click Analyze.
4. Review the mobile and desktop reports.
5. Examine the Core Web Vitals assessment and recommended improvements.

PageSpeed Insights combines two types of performance data: field data collected from real Chrome users and lab data generated through Lighthouse simulations.

The distinction matters. Field data tells you what visitors actually experienced over the previous 28 days. Lab data helps explain what might be causing a slow page.

The performance score runs from 0 to 100. A score of 90 or higher is considered good, but don't obsess over reaching 100. A high Lighthouse score does not guarantee good real-user performance.

For an e-commerce website, prioritize product pages, category pages, and checkout flows rather than testing only the homepage.

Google Lighthouse provides a deeper technical analysis of website performance, accessibility, SEO, and best practices.

You can run Lighthouse directly in Google Chrome:

1. Open the website in Chrome.
2. Right-click and select Inspect.
3. Open the Lighthouse tab in Chrome DevTools.
4. Select the categories you want to test.
5. Click Analyze page load to generate a report.

Lighthouse identifies issues such as render-blocking resources, oversized images, unused JavaScript, and long main-thread tasks.

It also reports laboratory metrics, including Largest Contentful Paint, Speed Index, and Total Blocking Time.

Important: Lighthouse does not directly measure Interaction to Next Paint (INP) during a normal page-load audit. INP requires user interactions. Instead, Lighthouse uses diagnostics such as Total Blocking Time to help identify potential responsiveness problems.

Use Lighthouse before deploying significant frontend changes and when investigating performance regressions.

The Chrome User Experience Report collects anonymized performance data from eligible Chrome users visiting public websites.

Unlike Lighthouse, CrUX reflects real-world conditions: different devices, connections, geographic locations, and user behavior.

CrUX provides field data for metrics such as:

• Largest Contentful Paint (LCP)
• Interaction to Next Paint (INP)
• Cumulative Layout Shift (CLS)
• First Contentful Paint (FCP)
• Time to First Byte (TTFB)

Google reports Core Web Vitals using the 75th percentile of real-user experiences.

In practical terms, if your website has a good LCP at the 75th percentile, at least 75% of measured experiences meet that threshold.

You can access CrUX data through PageSpeed Insights, the CrUX API, or Google's public datasets.

Some URLs do not have enough traffic to qualify for page-level field data. In those cases, PageSpeed Insights may display origin-level data or no field data.

While PageSpeed Insights tests individual URLs, the Core Web Vitals report in Google Search Console helps identify performance problems across groups of similar pages.

Open Google Search Console, select your property, and navigate to Experience → Core Web Vitals.

The report groups URLs by performance status and separates mobile and desktop data.

This makes it especially useful for larger websites. Instead of testing hundreds of individual product pages, you can identify groups of pages experiencing similar problems.

For example, if most product pages have poor INP, the problem may be shared JavaScript, a product configurator, or expensive variant-selection logic.

Fixing the shared component may improve hundreds of pages simultaneously.

Search Console uses CrUX field data, so improvements won't necessarily appear immediately after deployment.

Google's three Core Web Vitals focus on loading performance, responsiveness, and visual stability.

INP replaced First Input Delay (FID) in March 2024. FID measured only the delay before processing the first interaction. INP measures responsiveness across interactions throughout a page visit, making it a more representative indicator of usability.

You can learn more about performance architecture in our Frontend Performance Checklist, or explore the Crystallize documentation for implementation guidance.

Start with problems that affect real visitors, not whichever Lighthouse recommendation looks easiest to fix.

For ecommerce websites, the most common performance opportunities include reducing JavaScript execution, optimizing product images, improving caching, and avoiding unnecessary third-party scripts.

Large product images can delay LCP. Heavy JavaScript can hurt INP. Image galleries, personalization widgets, and promotional banners without reserved space can increase CLS.

The most effective workflow is straightforward: use Search Console to identify affected page groups, confirm the problem with CrUX, investigate it with Lighthouse, and measure the impact after deployment.

The goal isn't a perfect speed score. It's a website that feels fast to the people actually using it.