What’s your PageSpeed Insights score? Is it far from perfect? Is it three points away from 100? Causing you to rack your brain and surf the net incessantly in search of ways to get that sweet, sweet 100/100 score? Some of you in the MH Themes community are developers, so we know you take this score into consideration, even if a little. Some of you, however, are site owners with little to no technical inclination, so you rely on tools like PageSpeed Insights to ensure you’re doing things right.
PageSpeed Insights is a useful tool. It can help you pinpoint key technical issues occurring on your site that may have a negative impact on its speed and performance. However, the usefulness of this tool is not the question we’re asking in this post. The question we’re asking in this post is this: How important is a good performance score? More specifically, is a perfect PageSpeed Insights score a good representation of speed and performance?
- What is PageSpeed Insights?
- Why do we use PageSpeed Insights?
- The evidence
- The overall success of a website
- How important is a good performance score?
- How important is PageSpeed Insights?
What is PageSpeed Insights?
PageSpeed Insights is a development tool Google offers to let you analyze the performance of a website. In Google’s own words, it works by “fetching the URL twice, once with a mobile user-agent, and once with a desktop user-agent”. After it’s done analyzing the mobile and desktop versions of your site, it’ll give you a different score (between 0 and 100) for each. Google uses its own set of rules to analyze your site. These rules are separated into two categories:
You’re probably less familiar with its list of usability rules. These include configuring the viewport, adjusting the size of your content to fit the viewport and using legible font sizes. So, we’ve determined that speed and usability are the two groups of rules PageSpeed Insights uses to analyze your website and give it a score. This leads me to my next point.
Why do we use PageSpeed Insights?
We use PageSpeed Insights to gauge the performance of a site, specifically how fast it is and how well the user experience is handled. Unfortunately, that’s where the problem lies.
Every website ever built for any business has had the same main objective as the one before it, and that’s to earn the business more revenue. We know visitors aren’t going to stick around if they can’t load a website. We also know they’ll usually leave and never return if they can’t figure out how to navigate their way through a poor design. Therefore, user retention is one of the biggest reasons why speed (performance) and user experience matter.
Another major aspect is search engine optimization, or SEO as it’s better known. A site optimized for search has a higher likelihood of acquiring more organic traffic, one of the most valuable forms of traffic you can receive, than the other sites in its niche. Speed and user experience are both important ranking factors. So, does the score you receive from PageSpeed Insights indicate how fast your site is? How well it performs and how easy it is to use? No, and here’s why.
Do you remember how I said the problem lies in the fact that we use PageSpeed Insights to gauge the speed and user experience of a site? PageSpeed is full of theories but very little facts. It gives you a list of technical issues that may affect the speed and user experience of your site, but it doesn’t tell you whether or not those technical issues are actually affecting your site.
The biggest indicator of this? PageSpeed Insights doesn’t even list the amount of time it takes to load the pages you analyze. Let’s take a look at four different web pages to see what I mean. We’re going to compare the desktop score each page receives in PageSpeed Insights to the page load time it receives from Pingdom:
Let’s start with our own site. We’re going to test the MH Themes homepage. Here’s our PageSpeed Insights score.
And here are our results from Pingdom.
It’s pretty much what you expected, right? What happens when we test a website that has a poor score? Will that website load slowly or possibly even have a bad user experience?
Let’s test the mothership, WordPress.org. Here’s their PageSpeed Insights score, which may be a scary sight for some.
Here are their Pingdom results. You can see the site loaded in under 2 seconds.
Let’s test another.
The New York Times
The New York Times, founded in 1851, is one of the oldest news publications in the United States. Despite its age, the publication has had no trouble migrating and adapting to the modern web. Take a look at the PageSpeed Insights score of their homepage.
And here are their results from Pingdom. You can see their page load times are under the recommended 2-second limit used by many.
Let’s take a look at another news organization.
CNN, or Cable News Network, is a major news organization in the United States. It’s a lot younger than The New York Times as it was founded in 1980, but it was still founded before the web became mainstream. This is another example of an older company having no trouble migrating to the web. Take a look at its PageSpeed Insights score versus its page load times. Here’s PageSpeed’s score.
And here are their Pingdom results.
As you can see, a site’s PageSpeed Insights score is not enough of an indication of a site’s real-world speed (performance) and user experience.
The overall success of a website
This is another thing PageSpeed Insights can’t tell you. You can fine tune and tweak your site until all of those warnings go away and you get to as close to 100 as possible. However, what effect is that having on the site’s overall success? What I mean is, are those changes improving the speed and user experience enough to the point where they’re increasing conversions?
You can’t know these things without using third-party tools, such as Google Analytics or email marketing service providers. The overall goal of any online business is to increase revenue. However, as shown in the examples, your PageSpeed Insights score has no indication of the amount of traffic you’re able to attract and convert on your website. Which leads us right to the next question.
How important is a good performance score?
Okay, so we’ve determined that PageSpeed’s score usually is a poor representation of speed and performance, so let’s get back to our original question. How important is a good performance score? I think you already know my answer — not very. There’s a lot that affects a website’s speed, user experience and overall ability to convert visitors (into leads or customers). Here’s a few:
- Page speed.
- Quality of content.
- Content length.
- Content relevancy.
- Number of backlinks.
- Quality of backlinks.
- Anchor text.
- Email opt-in forms.
- Placement of opt-in forms.
- Use of relevant opt-in incentives.
All of these things are much more important than the score you receive from PageSpeed Insights. As a result, chasing the perfect score is only going to waste your time and resources or your client’s time and resources. Speaking of which, this is something to emphasize if you have a client that demands to see a 100/100 score in PageSpeed Insights.
As Sam Barnes wrote in an article published in Smashing Magazine, you need to speak your client’s language by considering their main business goals when you explain to them why focusing on that score usually isn’t a good investment. They’ll usually understand.
How important is PageSpeed Insights?
I just wanted to add this section to clear things up. This article had a very anti-PageSpeed Insights tone to it. I’m not suggesting for you to abandon the tool altogether. I’m only advising you to not focus your efforts on chasing a score that may get you nowhere in your business goals. Google themselves states this in their FAQs for the Insights tool:
Does the PageSpeed Insights score represent the real-world user experience?
PageSpeed Insights checks to see if a page has applied common performance best practices but does not account for all the factors that can affect a page’s speed – e.g. type of the device used to access the page, network speed, and cache reuse of shared resources. As a result, while the score is correlated with the speed of a page, it is not completely representative of the real world user experience.
It should instead be used as a diagnostic tool to help you determine what may be causing slow loading times, high bounce rates and low conversions. You should be cautious even then.
Check out this article by Andy Schaff over at Portent.com about how misleading and frustrating Google’s tool can be. He and a colleague implemented a responsive image solution that uses lazy loading and supports Google’s own webP image format on Portent.com. PageSpeed Insights didn’t detect it, but the PageSpeed Chrome extension did.
He also goes on to note the number of points Google may knock from your score for not setting expiry dates…for third-party applications. This is something that has to be done by the application itself. That means it’s something you usually don’t have control over. Therefore, it’s something that may always keep you from getting that perfect 100/100 score.
Final thoughts and suggestions
This is a controversial topic, and rightfully so. However, the conclusion is to simply take the metrics that have a direct indication of the overall health of your business into consideration over the score you see in the Google PageSpeed Insights tool. You should instead focus on optimizing your site for speed and user experience. We published an entire article on this if you need a bit of help: How to optimize the performance of a WordPress website.
It includes things like upgrading the hosting environment your site runs on, installing a caching plugin, image optimization and more. After that, you or your client could also optimize the way marketing is handled on the site, especially email and content marketing.
If you optimize all of this and still see issues with page load times, bounce rates and conversions, you may find a solution in PageSpeed’s suggestions. Bottom line, you should never implement a solution without knowing what the problem is. Do you have any questions or other helpful tips on optimizing the performance of a website? Please share your thoughts in the comments.