July 21st, 2021 at 8:45 pm
Is PageScore an Objective Metric for Grading Your Website Performance
19 minutes reading
What is PageScore and should I care about it?
Whether you are new to website building or an old-school webmaster, you have undoubtedly heard of the PageScore metric. There are loads of tools out there that help you measure it and ultimately help you optimize your website, making it faster. A fast website is a site that Google will rank higher in the search results and visitors will have a smooth browsing experience with, thus, making it their top choice when consuming the content you produce.
But what is common between website speed and PageScore? Are they related, and does having a 100/100 score will make your website fly? There is no ultimately correct answer, it is at the same time “yes” and “no”.
A higher score will unquestionably result in better performance. However, you do not need to be stressed or obsess over raising it. The only real metric which is essential is website speed! You could have a bad score(according to some tools), but your website can still load very fast. As we mentioned above, Google does not care about the score – it cares about Speed!
Common Webmaster Practices
So, what is the first thing you do when you create your website and host it somewhere? Of course, you go to a speed checking tool, such as PageSpeedInsights, GTmetrix, or PingdomTools. So far, so good.
Suddenly, you feel your head getting heavier, your pulse increasing, and a bitter taste in your mouth! The results are meager. The yellow, orange and red warnings are a lot more than the green ones! What happened here? At this point, you probably hate the world, you hate your hosting provider, and maybe even hate yourself a little bit.
We urge everyone who sees a low page score for their site to remain calm! Having low scores is not the end of the world! Ultimately, no search engine or visitor will care about the score, as long as the Speed of your website, its design, and its content are excellent.
If you are reading this article, and the above description is very familiar, please take a deep breath and continue reading, as we got you covered!
In this post, we will go over the most widely used website performance tools. We will explain which scores are relevant and will result in a considerable speed improvement.
We will also cover the ones that you should ignore. In 99% of cases, fixing those will require coding knowledge concerning HTML, CSS, JS, and PHP. Additionally, addressing them will improve your site’s Speed by a negligible margin and is not worth it.
The Myth about PageScore
Surely, you have read many articles out there and watched many YouTube videos of gurus and experts explaining the importance of having a high page score. Most of them are either made by developers who have a deep understanding of their website’s code and optimize sites for a living or by people who have no idea what they are talking about.
This polarity we are talking about can be confusing, and many people get bent on the idea of looking for that 100/100 score, and they eventually go nuts in the pursuit. Our advice – ignore those false claims and focus on delivering the best content, most fantastic design, and speed for your visitors.
You can associate PageScore with grading in school. Having lower grades than some of your classmates does not necessarily make you unclever, as many intelligent people have bad grades in school, whereas a lot of people cheat on tests and can get higher results – does this make them smarter?
Before we proceed further, let’s explain something about page loading speed.
Perceived versus Actual Loading Speed
You may have come across the terms Actual and Perceived loading speed. The Actual loading speed of a page is the time it takes for the browser to show everything on it – this includes videos, photos, external scripts, texts, and pretty much all of the content there is.
The perceived loading speed is the user’s perception of how fast they feel the website loading. Of course, this is a matter of perspective, as for someone, a site could load very quickly, whereas, for a different person, the Speed will seem slower. Naturally, performing optimizations on the Actual loading speed of a page will bring benefits to the Perceived Speed. However, it is much harder to pull off, as it will require development, technical, and financial means to achieve (in most cases).
Let us give you a vivid example that will explain what we mean.
Suppose you go to work and use the bus. Usually, the time it takes you to go to work is around an hour. To “speed” this process up, there are two actual things that need to happen:
- The Speed of the bus itself has to increase
- The distance to your job has to decrease
These two things are far harder to pull off, as they require a complete road infrastructure overhaul or a new bus with a brand new engine and more horsepower.
Now, the question is, how can you make this trip feel faster. It is pretty simple, really – what you need to do is make the journey more pleasant for you, and this will reduce your perceived time.
How can you do that? For instance, you can start carrying a book with you and read while you travel. You could also use your mobile data and browse around social media, listen to music, or watch a movie on your phone or tablet. Of course, any activities you do on the bus are up to personal preference. These activities affect your perceived time to get to work and make your experience a lot more entertaining.
Here are useful ideas of how you could apply the same concept for your website and improve the “Perceived” Speed of your pages:
- You could use the lazy-load tech for images if you have a lot of them on your page. This addition will gradually and progressively load them as the user scrolls down.
- Load the most significant content in the area “above the fold” on your page. This content is the first thing the user will see and will grab their attention while the rest of the resources are loading in the background.
- You can add progression indicators (loading bars, spinning circles, or any animation that signifies progression). You may find them online as “preloaders“.
These are just some of the standard “illusions” used by web designers and webmasters to conceal the actual loading speed and make the user experience more pleasant.
The problem with the tools that usually measure loading Speed and PageScore is that they show only Аctual Loading Speed results, and these little tricks that improve UX are considered bad practices, and the site will typically get bad grades in terms of Speed and size.
My WordPress is Slow How Can I Speed It Up?
If Speed is what you are truly after, then be sure to check our Speed Reaper hosting solution. We have integrated the famous LiteSpeed web server as a drop-in replacement for Apache. Not only does it serve content a lot quicker than its counterparts, but it also makes your website more stable, allowing it to handle far more traffic than it can on a typical Apache server.
It does not end there! As we offer LiteSpeed for our customers, they can integrate their websites with the LiteSpeed cache plugin, depending on their app. It blends perfectly with the webserver, taking the performance of every website to the next level!
Considering how famous WordPress has become in the last few years, we have invested the majority of our time and effort in fine-tuning our environment for WordPress and also provide our users the tools, know-how, and support they need to make it as fast as possible.
If you have a WordPress website that you think can use that extra speed it needs to outperform your niche competitors, be sure to take a look around in our blog’s optimization category. We have many articles that target the world’s most widely used application and more specifically – how you can improve its performance:
- WP-Rocket and it’s how it can sky-rocket your website
- Swift Performance – a free optimization plugin with a premium option
- Autoptimize & Cache-Enabler partnering to increase your website’s speed
If you are a customer of HostArmada and you wish to expand your website’s maximum potential, be sure to reach our technical support team over the ticket and request a review! Our experienced experts will surely provide great insight on how to speed up your WordPress and make you very happy!
Popular Tools to Measure Performance
We will go over three tools that we believe are widely used by website owners. We will explain how they work, their pros and cons, and which we recommend to measure performance. Note that you could use other tools available out there as well. Most of them work similarly and will show the same results. Of course, the practices we recommend in this post will apply to them too.
PageSpeed Insights is a popular tool by Google, which many webmasters use to weigh their website’s performance. Since November 2018, PSI uses LiteHouse technology and separates the analyzed data into two categories.
Field Data is a historical report of how your website performed throughout a given timeframe. It represents unbiased performance records from all the users who visit the website inside this timeframe from different devices and network conditions. This means that it may show faulty results in this category depending on the user’s ISP, their machine, and hardware. Additionally, if you have just built the website or transferred it over to a different hosting provider, note that this will change, and if you are getting bad scores now, you may see an improvement or decay in the upcoming months.
Lab Data, on the other hand, tests the page as it is right now, and it views a fixed set of conditions. Performing multiple tests in a row may show different results in that department. Most of the points it will make will be related to different types of resources on your site – JS, CSS, HTML, external resources, image sizes, encoding, etc.
Things to note when using PSI:
- You are likely to see a low score on mobile because the tool performs the test using a suppressed 3G connection.
- PSI does NOT show you how fast your website is.
- The location from which the tool performs the test is unknown. The location is a pretty big deal when doing a test, as it is natural that a website you host in the US will show worse results when visited in Asia.
- The device on which PSI performs the test is mid-tier and hardly represents mobile devices’ performance in general. Of course, people with older phones will have a worse experience than people using the latest flagships.
Despite the above, PageSpeed Insights gives you good advice regarding fundamental issues and can quickly fix them. For example:
- Properly scale your images: You need to make sure that you reduce your website’s images’ size as much as you can, without hindering their quality. This process will reduce the size of the page and, in that process, make it faster. If you are hosting your site with HostArmada and using WordPress, the LScache plugin offers a free option of scaling your images using quic.cloud!
- Server images in the next-gen format: Instead of using PNG or standard JPEG, you could use WebP for your image files, which offers excellent compression without sacrificing quality.
- Enable text compression: This will notify you if you are using a compression method for your static content. Any standard hosting server should provide either gzip or brothli.
- Server Response Time: This is an important metric you need to ensure is as low as possible. HostArmada offers a wide variety of data center locations that guarantee a fast server response time for your visitors!
- Add DNS prefetch: To enable this, you add the “preconnect” or “dns-prefetch” tag inside the files where you fetch third-party resources.
The rest of the recommendations are generally related to the site’s content that requires coding. It would be best if you ignored them unless you are familiar with programming code and have a deep understanding of the application you are using.
This tool is similar to the PageSpeed Insights, and it calculates identical metrics.
If you use this tool, you will notice two things that PSI lacks: the website load time indicator and the “Test from” option. If the “Load time” box gives you a value under 3 seconds, then you are excellent. Studies have shown that after 3 seconds of waiting, the user loses interest and is very likely to leave the website.
When selecting the location, you should pick the one closest to your targeted audience. If the results are not pleasing, consider switching to a data center near the visitors you want to attract. HostArmada offers 9 datacenter locations across the world, and you can pick whichever you want during the sign-up. If you are an existing customer, be sure to contact our technical team over a ticket and request a migration, which is COMPLETELY free of charge.
We recommend using this specific tool to measure performance as it gives you the most relevant information – loading speed!
A recent update to GTmetrix integrated Litehouse tech, and they use a very similar method to track website performance as PSI. They separated the scores into two categories: GTmetrix Grade and Web Vitals, which are practically the equivalent of “Field Data” and “Lab Data” we mentioned above. Instead of using Numbers, they use percentages from 1 to 100, which is generally the same.
Let’s go a quick rundown of the most important metrics you should look at in GTmetrix. We will use a website hosted with us, but we will not show the client’s domain for privacy’s sake.
- TTFB (Server Response Time) – The infamous TTFB(Time to First Byte), which you need to make sure is as low as possible because this is the ONLY metric Google and the rest of the search engines will consider when ranking the site. While this parameter does depend on the server where you host the site, in many cases it is hindered by the website’s content, even if the server itself is exceptionally optimized and polished. Another thing that can also affect the TTFB is the location from where you test the site. For instance, if you host the site in the US and perform a test from Asia, your application will have a higher TTFB. Additionally, the Server Response Time may be high under the following circumstances:
- Page Size: The content on your pages, such as images, text, scripts, ads, and videos, form the so-called “page size“. A well-optimized page should not exceed 1 – 1.5MB, as the larger the page is, the more time it will take to load (naturally)!
- Number of Requests: Each resource you have on the website (such as the one mentioned above) will be “called” by the server each time you visit your site. The more resources it has, the longer it will take the server to load them up and the browser to render them. A well-optimized page should not have more than 120-150 requests.
- First Content Paint – This shows your website’s time to generate images and text on the tested URL. For instance, if this value is 1 second – this means that it takes your website that much time to render any content on the screen.
- Time to Interactive – This performance metric shows how much time it takes your site to become interactive – in short, how soon can a client click around on links, posts, products, etc.
- Largest Contentful Paint – This metric will display how much time it takes for the most significant piece of content to become visible on your site.
- Onload Time – You can view this as the time when the browser has loaded the page entirely. For instance – if the site’s Onload Time appears to be very fast, you can consider this as the actual site loading speed. This brings us to the last thing you will see on the Summary page – The “Fully Loaded Time“.
- Fully Loaded Time – This is the final line you will see on the Summary Tab’s performance graph. Now, you should not panic if you see a big number here, as this shows the time in which EVERYTHING loads on the page – this includes ads, external links, scripts, and pretty much everything there is. Note that your website could be fully active a lot sooner, and the rest of the content, such as the one mentioned above, could load while the user is browsing. This fact is why Fully Loaded Time should not be considered a good indicator of user-perceived performance, and your focus should be on optimizing the site’s content, such as images, codes, themes, and plugins, first and foremost.
- Cumulative Layout Shift – This performance metric will measure your web elements’ unexpected shifting while the browser renders the page. The incremental layout shift depends solely on your website’s code, so we highly recommend specifying the width and height for the site’s media files. Additionally, you can use CSS aspect ratios to achieve the same result.
We believe GTmetrix is a better choice than PSI because it gives you an option to select the location from which to perform the test and includes information about your website’s Speed. You will see a bar separated into a couple of areas. The one you should pay the most attention to is the Largest Contentful Paint. This parameter will show the time it takes for the largest piece of content to become visible for your user. Naturally, the rest of the resources, such as images, videos, and various scripts, will load in the background while the user is browsing, so their experience stays intact.
All three performance testing tools we listed in this post are useful and could be a great asset when you want to speed up your website. It is crucial to understand how they work and which metric is relevant to performance so that you do not waste your precious time and nerves trying to solve a fictional problem. The tool that gives the most relevant information and the one we highly recommend to our readers is Pingdom. It gives a very clean, easily readable output containing the most valuable information – Page Loading Speed! Advanced users and web developers who have great knowledge of programming and can apply the proper changes to their content and code should use GTmetrix, and PSI to address what these tools recommend. We hope that this article helped you differentiate the good from the bad and you now have a clearer perception of what to look at when testing your website.