How do I quickly find out if my website is slow? Tools and a step-by-step guide

You can check your website speed in under 5 minutes with free tools. Open PageSpeed Insights by Google, paste your page URL and click Analyze. The result shows a score of 0-100 - a score below 50 means a serious performance problem, 90 and above is the goal. For every website, test the mobile and desktop versions separately, because Google primarily indexes the mobile version of a page. In this article we go through the exact procedure from the first test to concrete steps for fixing the most common problems.

If you haven't yet read what Core Web Vitals are and why they matter, I recommend starting there - you'll find precise definitions of LCP, INP and CLS without the jargon. This article assumes you know what you're measuring, and focuses on how to measure it and what to do with the results.

Why website speed matters for SEO, not just for the customer

Google has rated website speed as a direct ranking signal since 2021 through Core Web Vitals - a slow website therefore not only deters visitors, but directly drops in organic search.

In 2024, Google replaced the FID (First Input Delay) metric with INP (Interaction to Next Paint), further reinforcing the emphasis on page interactivity. This means an older audit from 2022 or 2023 may not reflect the current state of your website - INP is measured differently and an old site with good FID may now fail on INP.

Beyond SEO, speed has a direct impact on conversion. Mobile networks in Slovakia are relatively fast, but the processors in mobile devices are weaker than on desktop - a website must handle even a mid-range Android phone, not just a MacBook connected to Wi-Fi. A website that loads in 6 seconds on mobile is in practice unusable for most visitors.

Step 1: PageSpeed Insights - the first and fastest website speed test

PageSpeed Insights is a free tool by Google and the right start to any speed analysis - it provides both lab data and real field data from your website's visitors.

Available at pagespeed.web.dev. It combines two layers of data:

• Lab data (Lighthouse) - a synthetic test run in a controlled environment; always available, ideal for comparing before and after optimization
• Field data (CrUX) - real metrics from actual visitors to your website from the Chrome browser over the last 28 days; available only if the website receives enough traffic

If your website receives little traffic, CrUX data won't be available - in that case rely solely on lab data.

Procedure:

1. Open pagespeed.web.dev
2. Paste the URL of your main landing page (not necessarily the homepage - test the page that sells)
3. Click Analyze
4. Review the results separately for Mobile and separately for Desktop - always both

Score	Status	Practical impact
90-100	Green - good performance	The website is fast, CWV likely within norms
50-89	Orange - needs improvement	Visible slow spots, but not a disaster
0-49	Red - poor performance	Critical state, direct negative impact on SEO and conversion

Most Slovak presentation websites built on commercial templates with several page-builder plugins fall into the orange or red zone in mobile testing.

Step 2: Reading PageSpeed Insights results - what to look at first

From the PageSpeed Insights results, focus primarily on the Core Web Vitals values and the Diagnostics section - that's where you'll find specific problems sorted by their impact on performance.

Core Web Vitals thresholds according to web.dev/vitals (official Google documentation):

• LCP (Largest Contentful Paint) - when the main visual element of the page loads. Good: less than 2.5 seconds. Bad: more than 4 seconds.
• INP (Interaction to Next Paint) - how quickly the page responds to a click or key press. Good: less than 200 milliseconds. Bad: more than 500 milliseconds.
• CLS (Cumulative Layout Shift) - the degree of unwanted element shifting during loading. Good: less than 0.1. Bad: more than 0.25.

In the Diagnostics section, look mainly for:

• Render-blocking resources - JavaScript or CSS files that block the page from displaying before loading completes
• Unused JavaScript / Unused CSS - code that is downloaded but not used; typical for websites with page builders
• Images not properly sized - images larger than the website actually displays; they waste bandwidth unnecessarily
• Large network payloads - the total size of the page; roughly under 2 MB is healthy

The first two problems - render-blocking and unused JavaScript - are typically the biggest issue for websites built on Elementor, Divi or WPBakery. We wrote about why and how to get out of such a setup in the article Migrating from Elementor: how to switch to a clean WordPress theme.

Step 3: GTmetrix for deeper diagnostics - where time is lost

GTmetrix offers a more detailed view than PageSpeed Insights, including a waterfall chart and a test from European servers - which gives a more realistic picture for Slovak visitors.

A free account at gtmetrix.com lets you test from various geographic locations. This matters: PageSpeed Insights tests from servers in the USA. A website hosted on an SK or CZ server will have different results when tested from Europe versus America. GTmetrix with a European server (London or Vancouver - choose London for SK relevance) gives a more realistic picture.

The key view is the Waterfall tab - a visual representation of the order in which and how long each file takes to load. If you see one long bar (a JS bundle, massive CSS or a large image), that's where the bottleneck is.

GTmetrix also shows:

• The total page load time (Fully Loaded Time)
• The total volume of transferred data in MB
• The total number of HTTP requests

Rough benchmarks for a well-optimized website on SK/CZ hosting: load time under 2 seconds from a European test server, total page size under 2 MB, number of requests under 80. These values are only indicative - they depend on the complexity of the page and its content.

Step 4: Google Search Console - data from your website's real visitors

Google Search Console contains a Core Web Vitals section with real data from your visitors - this is a more authoritative source than synthetic tests, because Google evaluates exactly this field data.

If you have Google Search Console set up (and you should), find the Core Web Vitals section in the left-hand menu. The report divides URLs into three groups:

• Good URLs - all three metrics in green
• URLs that need improvement - some metric in orange
• Poor URLs - some metric in red

The Search Console CWV report is more important than the Lighthouse score, because for ranking Google evaluates field data, not lab data. If your website receives at least a few hundred monthly visits, Search Console is the most authoritative source for assessing performance.

Search Console also links performance problems to specific URLs - you can see whether the entire domain is slow or just specific pages (e.g. a blog category vs. the homepage). That makes prioritizing fixes much easier.

Comparison of website speed testing tools

Each tool has a different angle - for a complete picture it's ideal to combine at least PageSpeed Insights with Search Console.

Tool	Price	Data type	Best for
PageSpeed Insights	Free	Lab + Field (CrUX)	First test, CWV metrics, instant result
GTmetrix	Free / paid	Lab (synthetic)	Waterfall, test from a European server, comparison over time
Google Search Console	Free (requires verification)	Field (real users)	Production data, SEO context, identifying problem URLs
Chrome Lighthouse (DevTools)	Free	Lab	Developer diagnostics, testing changes before deployment
WebPageTest	Free	Lab (advanced)	Advanced analysis, film strip, frame-by-frame testing

Typical problems of Slovak websites and how to fix them

When analyzing Slovak presentation websites, we repeatedly run into the same five problems - most of them are solvable without completely rebuilding the website.

1. Unoptimized images
Photos at their original size, JPEG without compression, PNG where WebP would suffice. A 4-5 MB image on the homepage is a critical problem on a mobile connection.
Solution: Convert to WebP or AVIF format, set lazy loading for images below the "fold", add width/height attributes to prevent CLS.

2. Page-builder dependency
Websites on Elementor, Divi or WPBakery load dozens of JS and CSS files on every page. Many of them are render-blocking and can't be easily removed without changing the architecture.
Solution: Migrating to a clean WordPress theme without a page builder is radical, but effective. If you're seeing 5 or more signs that your website is falling behind, read 5 signs it's time for a website redesign.

3. Slow hosting (high TTFB)
A Time to First Byte (TTFB) above 600 ms is a hosting or server configuration problem, not a front-end one. No front-end optimization will erase a slow server.
Solution: Move to hosting with SSD drives, PHP 8.2+ and LiteSpeed or Nginx available. For WordPress: LiteSpeed Cache or WP Rocket combined with a CDN help significantly.

4. Unconfigured cache
Every visit re-downloads the same CSS, JS and images. Proper cache configuration saves a significant percentage of transfers for returning visitors.
Solution: Enable server-side caching (LiteSpeed, Nginx FastCGI), set browser cache headers and consider a CDN for static assets.

5. Uncontrolled third-party scripts
Google Analytics, Hotjar, Facebook Pixel, a chat widget, a newsletter form - each script adds latency and may be render-blocking. On some websites we see 8-12 third-party scripts without the owner knowing why they're there.
Solution: Audit all scripts, keep only the essential ones, remove the unnecessary ones. Load the rest asynchronously or with the defer attribute.

In practice: how to follow up on the test results

The test alone isn't enough - the results need to be turned into a concrete list of steps sorted by their impact on Core Web Vitals.

In practice, when analyzing Slovak websites we run into the same pattern: the owner had the website built 4-5 years ago, and back then it worked well. Today Google evaluates INP (a new metric since 2024), mobile indexing is a priority and the page loads in 6-8 seconds on a common Android phone. The result: a higher bounce rate, no organic leads.

The good news: most problems are solvable without completely rebuilding the website. Prioritize based on which fix has the biggest impact on CWV:

1. LCP problem = usually an image or server response time - fix this first
2. CLS problem = missing image dimensions or web fonts without size-adjust - quickly fixable
3. INP problem = too much JS on the main thread - more complex, requires a JS audit

If the results show red or orange metrics and you don't know where to start, a comprehensive SEO and performance audit helps - it identifies the problems and assigns them priority. Websites we build from the ground up without click-together templates start with significantly better performance - unnecessary code simply doesn't get into them.

In the context of Google AI Mode and LLM citations (we wrote about it in Google AI Mode and SEO 2026) page performance is even more important - slow pages have a lower chance of being cited as sources in AI answers, because Google rates them lower on Page Experience signals as well.

Conclusion

Testing your website speed takes 5 minutes and the tools are free. PageSpeed Insights and GTmetrix give a clear picture of the state of things, and Google Search Console then shows how real visitors see it. If the results show red numbers and you don't know what to do with them, take a look at what our SEO and performance optimization covers or - if the problem lies deeper in the website's architecture - a website redesign.

Thinking about a performance audit of your website? A no-obligation consultation - we'll reply within 24 hours.

FAQ

Frequently asked questions.

What is a good score in PageSpeed Insights?

A score of 90 and above is green - the website is fast and Core Web Vitals are likely within norms. A score of 50-89 is orange and signals visible performance problems. Below 50 is red - the website has a serious problem that directly affects both SEO and conversion.

Why does my website have a different score on mobile and on desktop?

Google evaluates mobile and desktop separately, because mobile devices have a weaker processor and slower networks. The mobile score is usually lower and more important - Google primarily indexes the mobile version of a page (mobile-first indexing).

Does it matter which location I test the website from?

Yes. PageSpeed Insights tests from servers in the USA - for an SK website on European hosting, GTmetrix with a European server (e.g. London) gives a more realistic picture. We recommend testing with both tools.

Does speeding up a website help SEO too?

Yes - Google has used Core Web Vitals as a direct ranking signal since 2021. A slow website gets lower positions than a content-comparable fast website. More about the metrics in the article on Core Web Vitals.

What is TTFB and why does it matter?

Time to First Byte (TTFB) is the time from the first request to the first byte of the response from the server. A TTFB above 600 ms indicates a hosting or server problem. No front-end optimization will fix it - you need to address the hosting, server configuration or CDN.