Website speed test. Where and how to measure it? Every millisecond counts

Rychlost webu je jedním z důležitých parametrů, které ovlivňují Vaši pozici na search engines Google. And not only that. Your website’s speed affects how satisfied visitors are with your website: on a faster website, people buy more, stay longer and read more.

Milliseconds matter

The big players providing global data services put their heads together and carried out in-depth research. The result is a list of facts showing why milliseconds matter:

• For every 100 milliseconds of additional page load time, the internet giant Amazon saw a 1% drop in sales (Kohavi and Longbotham 2007)
• Google found that when it changed its website to show 30 search results in 0.9 seconds instead of 10 search results in 0.4 seconds, traffic and revenue fell by 20% (Linden 2006)
• Google also found that a 400-millisecond (0.4-second) delay in displaying search results led to -0.59% searches per user. What is more, when Google sped search back up by 400 milliseconds for the same test group, they still made -0.21% fewer searches. Google thus discovered that website speed affects the long-term behaviour of your website’s users.
• Yahoo (another search engine, not used in the Czech Republic) also found that on a page 400 ms slower, 5 to 9% more users close the page before it loads.

And so on. Simply put: Speed matters.

You can measure your website’s speed from different corners of the world for free on this website

http://cloudmonitor.ca.com/en/checkit.php (enter your website’s address and click Check)

What a website speed test actually measures

A website speed test loads your page in a controlled environment and records how long each part takes to appear. You get several measurements, each describing a different moment of loading.

• Time to first byte shows how quickly your server starts sending the page after the browser requests it. A slow value usually points to hosting, server configuration or a heavy back end.
• Largest Contentful Paint shows when the main visible element has loaded, such as a large image or the main heading. This is close to the moment visitors feel the page is ready.
• Interaction to Next Paint shows how quickly the page reacts when someone clicks, taps or types.
• Cumulative Layout Shift shows how much the content jumps around while the page loads.
• Page size and number of requests show how much data the browser has to download and from how many files.

Tools also differ in the data they use. Lab data comes from one simulated visit with a fixed device and connection. Field data is collected from real visitors over a longer period. Lab data helps you find the causes of problems. Field data shows what people actually experience.

Other tools for measuring your website’s load speed and identifying other problems

• https://developers.google.com/speed/pagespeed/insights/ A tool straight from Google, the world’s largest search engine. It checks your website and lists tips on what you could improve. If you cannot manage it yourself, try discussing it with a developer.
• http://tools.pingdom.com/fpt/ A very nice service for measuring your website’s speed, although only from one location.
• http://gtmetrix.com/ Another tool for measuring your website’s load speed.
• It is also handy to monitor your website’s availability around the clock. We wrote about the free Uptime Robot service.

Do not underestimate your website’s load speed. Here is the measurement result for this website (WebyChytre.cz):

Mobile website speed test and common mistakes

Many visitors browse on phones, which often have slower processors and less stable connections than a desktop computer. A page that feels fast on your office computer can feel sluggish on a mid-range phone. Most free tools simulate a mobile device on a slowed-down connection, so the mobile score is usually lower than the desktop one. This is normal. Your aim is to improve it, not to make it match the desktop score.

Mistakes to avoid

• Chasing a perfect score. What matters is that your real visitors get a page that loads and responds quickly.
• Relying on your own browser. Your cache and fast office connection hide problems that new visitors notice.
• Comparing numbers from different tools. Each tool uses its own settings, so compare each tool only with its own earlier results.
• Ignoring third-party scripts. Chat widgets, tracking codes and embedded videos often slow a page more than your own code.
• Uploading oversized images. Large photos that the browser scales down are a frequent cause of slow pages.

How to test website speed step by step

• Choose the pages that matter. Test the homepage, but also a product page, a blog article and a page with a form.
• Pick a test location close to your audience. If most of your customers are in Central Europe, a test server on another continent will show slower results than they experience.
• Run the test several times. Results vary with network conditions and caching, so look at the typical result rather than a single run.
• Test mobile and desktop separately. They often reveal different problems.
• Read the waterfall chart. It lists every file in the order it loads, so you can see which image, script or font holds the page back.
• Fix one thing at a time and test again. This way you know which change helped.
• Repeat regularly. Test again after each larger change, such as a new plugin, theme or tracking code.

Frequently asked questions about website speed tests

Is there a free website speed test?

Yes. PageSpeed Insights, Pingdom and GTmetrix, mentioned above, all offer free tests. Paid plans usually add monitoring, more test locations or saved history, but a free test is enough to find the main problems.

How do I test website speed on mobile?

Use a tool that offers a mobile test, such as PageSpeed Insights, which simulates a phone on a slower connection. You can also open your website on an ordinary phone using mobile data and watch how it behaves. Both views together give you a realistic picture.

What are the best website speed test tools?

No single tool is best for everything. PageSpeed Insights combines lab and real-visitor data, while GTmetrix and Pingdom show detailed waterfall charts. Using two of them together usually gives you the clearest picture.

Why does my website show different results in each test?

Results depend on the test location, the simulated device, the connection speed and whether files were cached. Server load at the moment of testing also plays a role. Run several tests and compare trends rather than single numbers.