This website was recently added to the CrUX report, meaning you can now see this websites real-world performance, because the CrUX report shows data from Real User Monitoring (RUM).

Real-world performance metrics for usecue.com

RUM differs from synthetic monitoring in that it relies on actual people clicking on the page to take measurements rather than automated tests going over a given set of steps, like a standard Google Lighthouse test.

It is much harder to get a perfect Google Lighthouse score with RUM than with synthetic monitoring, because in the latter you control all variables. When we analyze the RUM results (click the picture above), we can see a few things.

First of all, you can see a ‘high’ Time To First Byte of 1.1 seconds. This is due to the fact that this website does not use a CDN, while it has a lot of visitors from Asia and North America and it is hosted in Europe. Therefore, most traffic has to travel more than 9 thousand kilometers.

The second thing we see is that the site (on average) takes another 0.5 seconds to get to the FCP and LCP. This makes the total (experienced) page load 1.6 seconds. This is not a great score, but it is not bad either.

We know that this website can load in 0.1 seconds in a synthetic test. It requires a fast computer and a optimal geographical location. But… even without a CDN, when most traffic has to travel to the other side of the world and when opened on real-world devices, we can STILL get a perfect Google Lighthouse score on desktop in Google’s RUM report. All scores are well within the green areas. Isn’t that amazing?

I have a lot of Japanese readers. If they could experience the extreme speed Dutch visitors are experiencing, they would be blown away. I could add a CDN with a node in Japan to fix this, but it would slow down the speed in the Netherlands, as all CDN’s have overhead. I deployed my website to Cloudflare (today) and it instantly became 4 times slower (load time increased from 110ms to 440ms). Also, some Japanese people already think I have the fastest website in the world. So, should I slow down my website for Europeans to make it faster for Japanese people? I think not. Finally… I like to show to the world how fast a single node website (on proper hosting) can be. I am also not very interested in measuring the speed of Cloudflare’s (inferior) CDN.

()  Joost van der Schee