Semalt's Google PageSpeed Guide For WordPress Users 

Google PageSpeed Insights (PSI for short) is a nice tool for web developers. Its intentions are certainly good, but it can be said that it is not suitable for the average WordPress site owner. And although PageSpeed Insights recently introduced specific comments for WordPress users, many aspects of the report it provides are too technical for the average users.

In this guide, we will try to "translate" and describe in simple words what PageSpeed is talking about and we will explain to you which comments should be addressed and treated and which should not. This part of the guide will focus only on understanding and analyzing lab data.  However, in another article, we will talk about the same suggestions and opportunities for improvement that PSI offers.

The basic problem that PageSpeed Insights is trying to eradicate

In recent times the popularity of WordPress sites and other content management systems has grown miraculously. These allow users with minimal knowledge of development, or those with no knowledge of the subject at all, to build sites whose way of building contradicts the basic rules of site performance.

Those Page Builders that allow anyone with no knowledge in website development have a significant impact. The fact is that if you build a simple and lightweight website, it will automatically get a higher score in these tools, and more importantly, it will be much faster for the surfers and hence the user experience will be better.

Google is trying to encourage a good user experience and eradicate these issues, among other things through a project you've probably heard of called AMP. Google says "company, if you add junk to your site, then at least let's see what can be done to reduce its impact on speed and loading time."

Beyond that, Google - whose business model is advertising and services - must make sure that the speed of sites and the user experience on the web will improve. Otherwise, closed platforms such as Facebook, which also offer advertisements, will eat up an even larger share of its revenue.

How do I understand the information that Google PageSpeed displays?

Google's speed testing tool is currently based on Lighthouse and evaluates each site based on a set of rules. These rules apply to each site regardless of its target audience, its content, or its purpose.

PageSpeed Insights provides reports on site performance on mobile and desktop and provides information on opportunities and suggestions on how to improve the performance of the page under review.

Area data versus laboratory data (test data)

PSI provides lab data as well as field data. Only sites with particularly high traffic will receive information in a section called area data so there is no need to worry if you do not receive it.

We will not expand on field data in the current post, but it is said that the laboratory data, which we will talk about, are collected in a controlled environment using predefined devices and fixed network settings.

Laboratory data are the ones that are relevant in favor of problem monitoring, debugging, and optimization because these conceptually do not change. Laboratory data provide more information than area data, however, these do not provide bottleneck information that can be generated in the real world and affect site performance.

How to use Google PageSpeed?

Before we begin, it is said that the first step to making sense of the report presented by Lighthouse is to know that you have to click on the arrow that appears next to each comment to get detailed information about it.

You need to dig deeper to understand and see the problems the report presents and I suppose that is understandable. Let's start with the overall score that PSI provides I'm pretty sure most of you do not know exactly what this score reflects…

Overall score
PageSpeed - General score: The overall score provided by PageSpeed summarizes page performance by number and color. The score is determined by running a Lighthouse and provides so-called "sterile" lab data on the same page being tested.

One of the problems with PageSpeed is displaying the score in some color and number that not everyone can understand. Green color = good, red color = not good. The use of this method and the lack of understanding of the meaning of the overall score make people frustrated that their site performance is "supposedly" low and they have no idea exactly how to fix it.

Because most users think that something less than green is not good enough, people misinterpret the report. So let's get things in order once and for all…

How to interpret the overall score
If you do get green, it means that your site is in the top percentage in terms of optimization and it does not say much about the actual site speed.

A score of 100 is the best score you can get and represents the 98th percentile - a site with (apparently) good performance. A score of 50 represents the 75th percentile - that means your site received a score higher than 75% of the sites that Google checked and it does not say anything about the actual speed.

A score of 90 or higher is considered a good score, 50-90 is considered average, and below 50 is considered low.

Hence, most sites will not get a green score. An average WordPress site built by a site builder with average technical capabilities will probably not get a high score Out of the box. And despite optimizing them, these sites usually have a lot of elements that harm the score to obtain.

What is PSI used for?
PSI provides feedback and opportunities for improvement to help you fix issues that have this negative impact. The same "test data" (laboratory data), which we will talk about soon, provide up-to-date information in terms of schedules.

For example, when your page is displayed to the surfer and when the surfer can start performing actions on the same page. These are specific measurements that affect the overall score.

Lab Data - Lighthouse

If you plan to run a Lighthouse test through Chrome Developer Tools - make sure you do so in a window where no browser extension is enabled.

Let's expand on those lab data and describe them one by one. In general, your goal in the context of this data is to make each as short as possible:

1. Presenting the initial content - First Contentful Paint
Initial Content Display (FCP) is a measurement of the time from the moment the page is requested to the time the browser renders the first bit in the DOM. This is an important milestone for the user because it provides feedback for the page to load.

In other words, the browser must process the page code to display it. The term "Paint" refers to the process of actually displaying the pixels on the screen. The initial content display figure describes how long it took for any content, whether text, image, or any other content to be displayed on the screen for the surfer.

2. Presenting the main content - First Meaningful Paint
The Main Content Display (FMP) figure tries to identify the time when the surfer feels that the main content of the page is displayed for him. Of course, this type of test cannot know exactly what the main content is and therefore is not accurate.

In general, Lighthouse defines this as the time when the most significant change is made to the critical part of the page (Above the fold content) in addition to the time when the web fonts are loaded.

3. Speed Index
The speed index is quite similar to the two previous measurements we described. It generally shows how fast the page content is displayed to the surfer and is also weighted according to the same previous measurements we mentioned. Of course, a lower score = better.

4. First CPU Idle Mode - First CPU Idle
This figure (FCI) and the next figure called "Time to Interactivity" refer to the speed at which the user can perform actions on the page in question. If he tries to perform an action while the browser is processing the code, it is said to be a button press. Then either he will have to wait quite a bit until the action is performed or the action is not performed at all. 

The CPU inactivity figure indicates how long it takes for the page to be interactive, i.e. to reach a state where the user can interact with most of the elements on the page and get a reasonable response time.

5. Time to Interactive 
The Time To Interactivity (TTI) figure measures how long it takes a page to be fully interactive when interacting with elements on the same page such as a click of a button to get a quick response (about 50ms).

Page interactivity is measured by several factors:
  • The page displays content relevant to the user (weighted according to the data displayed in the initial content).
  • Event Handlers are listed for most of the visible elements.
  • The page responds to user actions within about 50ms.
There are quite a few sites that optimize so that the content is displayed faster but at the expense of interactivity, which can create frustration for the user. The site seems to be fully loaded but when the user tries to perform any action nothing happens.

6. Maximum Potential First Input Delay
The maximum first input delay figure is related to the length of time it takes for the browser to perform the longest operation, for example, this will delay the response speed of a button click by the user.

For example, in case the browser is busy processing a certain JavaScript on the page (which is not asynchronous), you will probably not be able to perform certain operations until the processing is complete.

In conclusion

In this part of the guide, we talked about the lab data that PageSpeed provides us and explained the meaning of each of these data. As promised soon we will talk in detail about each of the opportunities or comments that PageSpeed Insights presents to us.

Last words -Do not go crazy to get the highest scores, make sure that the loading time of your site pages is reasonable + it is not unusually slow so that the impact on the user experience is not significant. Remember that a score of 50 means that the optimization status of your site is better than 75% of the sites reviewed.

And once you get to this point, focus on quality content and invest in the other parameters included in the technical SEO. If you do not have the necessary knowledge in the field of reference, then contact the experts of Semalt. You will have excellent results in a short time thanks to the autoSEO pack.

This SEO pack is a Real "full house" for an online company that aims to:
  • Improve the visibility of your website
  • Optimize each page of your site for better conversion rates
  • Building quality links
  • Search for relevant keywords
  • Make a detailed web analytics report
 If all this is done well on your site believe me that even if your site's PageSpeed score is lower than that, you will end up ranked well... then good luck!

send email