FAQ

Our vision is to enhance the user experience of websites. We believe that speed has a major impact and that improving it helps businesses achieve their website goals more effectively.

We provide web performance consulting and understand that the tools people use don’t always serve them well.

Our speed tester enables businesses to monitor performance regularly, preventing unexpected slowdowns.

We operate two versions of the speed tester, and the method of measuring pages varies accordingly:

Free Version

We retrieve website speed data from the PageSpeed Insights API, always during nighttime hours via Amazon Web Services. Desktop is tested every two days, and mobile is tested every two days as well.

This data is freely available but unstable since Google does not guarantee a consistent testing environment.

The PageSpeed Insights API measures mobile performance using an emulated Moto G4 device on a network with roughly 4G speeds. Desktop testing is conducted on a 10 Mbps connection. The tester's location is in Europe.

PLUS Version

For the paid version of the tester, we have built our own testing infrastructure, which runs each URL through Lighthouse on Amazon Web Services from Frankfurt, Germany. As a result, the data is more stable and better reflects real-world performance.

We also conduct tests more frequently. Every URL is tested daily, five times in a row, to minimize extremes and fluctuations, both on mobile and desktop.

With the PLUS version, you also receive more detailed technical data, including a Trace export of the page rendering process, which can be opened in Chrome DevTools.

Our infrastructure tests using an emulated Moto G Power device, running Lighthouse 11.1.0 and Chrome 114..

Chrome UX Report

In both versions, we also fetch user data from Google's Chrome UX Report for both domains and specific URLs.

We retrieve monthly data once a month and rolling 28-day data for URLs and domains on a daily basis.

There can be several reasons:

1. The test is new, and data takes some time to appear. For free tests, data usually appears within a few minutes. Data retrieval for domains from the Chrome UX Report is done once a month, so it may take longer.
2. No data is available for the given URL or domain. If you don’t see numbers for Google user data, it may mean that the data is unavailable. Google provides Web Vitals and other metrics only for pages with a certain level of traffic. This can also happen with synthetic data if the page blocks bots in some way.
3. The page or domain is new. Google provides cumulative data from the last 28 days, meaning new websites will receive data with a delay.

If you believe you should be seeing data despite the reasons above, please contact us at info@pagespeed.cz or via chat.

There are various reasons for different measurement history lengths:

1. If the test is not part of your dashboard, you will only see a monthly history. Logged-in dashboard users can view three months of data. In paid PLUS tests, you will see a full year of history.
2. Some URLs may have been tested by someone before you. In that case, we have good news – the test will display the entire history of those measurements. This means different addresses in your test may have varying history lengths. We simply didn’t want to deprive you of valuable data from before you created this speed test.
3. Conversely, URLs from tests that haven't been viewed for more than three months will no longer be measured.

The most important metrics are the Core Web Vitals collected from real users via the Chrome UX Report (CrUX):

• LCP (Largest Contentful Paint) – loading speed. The ideal value is under 2.5 seconds.
• CLS (Cumulative Layout Shift) – layout stability. The ideal value is under 0.1.
• INP (Interaction to Next Paint) – interaction responsiveness. The ideal value is under 0.2 seconds.

The INP metric has been in effect since March 2024, replacing the older and less suitable FID.

The table displays the 75th percentile for the set of measurements. More details on this percentile selection can be found in the measurement API documentation.

The ranking of metric importance from different sources is as follows:

• User-based metrics (CrUX) for the domain.
• User-based metrics (CrUX) for a specific URL.
• Lighthouse metrics for a specific URL.

This is completely normal, and the reasons may include the following:

• Different data sources (synthetic or real users).
• Different data collection methodologies (varied Lighthouse measurement settings).
• Measurements taken at different times, influenced by your website's load or the display of various ads.

It’s important to measure data from stable sources and user-based metrics (CrUX data) and to use a single tool for long-term tracking. That’s why we developed our tester.

The PageSpeed.cz project was born from the long-term collaboration between Martin Michálek and SUPERKODERS.

Have any other questions? Feel free to reach out at info@pagespeed.cz or via chat.