PageSpeed Secrets: Why doesn't PageSpeed measure the Loading Time?
Google PageSpeed Insights Secrets
Did you know that Google PageSpeed doesn't measure the loading time?
Are you one of the thousands of users who have always believed that Google PageSpeed Insights is about load time? There may be many reasons for this very common misconception, but you too have probably been misled by the word speed in the name PageSpeed? Especially since what else should it be about than the loading time?First of all, let's note that Google PageSpeed is not about loading time. You can find out what PageSpeed is actually about in the Google PageSpeed Insights Secrets.
Google PageSpeed is not about the loading time, but about the so-called display time. There is a huge difference between load time and display time, but both have something to do with speed. You have to keep in mind that measuring the loading time depends on many factors and is influenced by just as many factors. PageSpeed, or Google operates a scoring and you will certainly agree that scoring not only has to be fair, but that every website has to be checked according to the same criteria. The problem, however, is that it is simply not technically possible to test every page under the same test conditions if you want to measure the loading time, because above all the distance between 2 servers can never be the same and the performance of servers varies greatly. Otherwise, the servers that are closer to a Google test server and more powerful than a more distant shared hosting server would be favoured.
That sounds logical, doesn't it?
And precisely for this reason, Google cannot and does not want to measure the loading time. But what is the difference between loading time and display time and what does Google actually measure?
The loading time describes the time it takes to load all sources from a server. However, loading a source from a server does not mean that something is also displayed in the browser. The loaded data must first be processed by the browser. So when we talk about display time, it starts from the moment all data has been loaded and only then does PageSpeed start measuring. The actual loading is almost completely ignored when measuring and scoring. For Google, the only interesting thing is how quickly the data is processed, because only that can be measured almost unaffectedly using the same criteria.
That sounds logical too, doesn't it?
If everything can now be explained logically, one would actually have to blame Google that PageSpeed is not about the speed of a page at all. At least the name PageSpeed is misleading and the score says nothing about the speed of a page. This is especially true when differentiating between an XXL high-end dedicated server and a 5 dollar shared hosting. A dedicated server is always faster and more powerful than low budget shared hosting. But despite this difference, Google punishes owners of a dedicated server with a bad score and pretends that shared hosting operators have a fast website.
What's missing from this apparent injustice is that Google only focuses on the level of optimization. However, this important aspect not published. The actual loading is completely left out. Ultimately, however, it is the case that a server, no matter how powerful but insufficiently optimized, which can provide the data much faster, is rated worse according to the PageSpeed criteria than a maximally optimized but terribly slow page on shared hosting.
What Google does with PageSpeed is therefore highly misleading. Especially users of shared hosting are affected because, as already mentioned, they are led to believe that they have a fast website because the site is well optimized. In fact, an optimized webpage is only optimized for display time, but not for loading. No matter how good optimization is, a website will not become faster. This only improves the display behavior. This changes almost nothing in terms of the amount of data to be loaded. Likewise, optimizing does not change the source loading time.
It would therefore be wrong to think that one is rewarded for working on optimization. Although Google promises a better placement in the search engines with a high score, nobody knows what influence a good score actually has. Above all, these deceived users lose focus on what is actually important and that is not Google, but the users who visit a website. What use is a score, no matter how good, if the website loads slowly and the visitors jump off prematurely as a result or the dwell time is only short because of the long loading time?
We can therefore only advise anyone who is trying to create a fast website to focus on the visitor and not on Google. But that doesn't mean you don't need to optimize your website. For us, optimization means a holistic approach and this includes optimizing the page according to the criteria of PageSpeed, but also optimizing the loading time so that the main document in particular loads faster. This is because as long as the main document is not loaded, no static sources such as CSS and Javascript are loaded either.
Optimizing the loading time is not only achieved through more powerful hosting. With an HTTP cache such as the LiteSpeed Cache, it is easily possible to outperform a dedicated server with inexpensive shared hosting many times over. Therefore, a fast loading time is not dependent on the available budget. Above all, LiteSpeed hosting with LScache does not cost more than conventional hosting with Apache or nginx.
What's missing from this apparent injustice is that Google only focuses on the level of optimization. However, this important aspect not published. The actual loading is completely left out. Ultimately, however, it is the case that a server, no matter how powerful but insufficiently optimized, which can provide the data much faster, is rated worse according to the PageSpeed criteria than a maximally optimized but terribly slow page on shared hosting.
What Google does with PageSpeed is therefore highly misleading. Especially users of shared hosting are affected because, as already mentioned, they are led to believe that they have a fast website because the site is well optimized. In fact, an optimized webpage is only optimized for display time, but not for loading. No matter how good optimization is, a website will not become faster. This only improves the display behavior. This changes almost nothing in terms of the amount of data to be loaded. Likewise, optimizing does not change the source loading time.
It would therefore be wrong to think that one is rewarded for working on optimization. Although Google promises a better placement in the search engines with a high score, nobody knows what influence a good score actually has. Above all, these deceived users lose focus on what is actually important and that is not Google, but the users who visit a website. What use is a score, no matter how good, if the website loads slowly and the visitors jump off prematurely as a result or the dwell time is only short because of the long loading time?
We can therefore only advise anyone who is trying to create a fast website to focus on the visitor and not on Google. But that doesn't mean you don't need to optimize your website. For us, optimization means a holistic approach and this includes optimizing the page according to the criteria of PageSpeed, but also optimizing the loading time so that the main document in particular loads faster. This is because as long as the main document is not loaded, no static sources such as CSS and Javascript are loaded either.
Optimizing the loading time is not only achieved through more powerful hosting. With an HTTP cache such as the LiteSpeed Cache, it is easily possible to outperform a dedicated server with inexpensive shared hosting many times over. Therefore, a fast loading time is not dependent on the available budget. Above all, LiteSpeed hosting with LScache does not cost more than conventional hosting with Apache or nginx.