After the creation of a website and a valid SEO optimization , it is good practice to carry out tests . Put yourself in the end user’s shoes, browse the site in its entirety and try to carry out tasks. For example: submit a form, complete a purchase or download a PDF. In short, you need to check the contents, test the functioning of the links and verify that the pages load correctly .
Let’s stop on the last point. When we talk about website speed we always end up on Google PageSpeed Insights . An undoubtedly powerful and useful tool, but can we really trust it with our eyes closed and take every judgment it makes seriously?
After carrying out some tests and scouring the internet on this topic, we came to a rather surprising conclusion. Let’s discover it together.
Evaluating loading speed: the (wrong) approach
Loading speed is a crucial aspect for every website. Both for users and for search engines . Both want content – of quality – that is accessible and usable in the shortest time possible . If a page doesn’t load immediately, the offending session turns into an inevitable bounce.
During the testing phase, a question we usually ask ourselves is “ Is the site fast? “. A more than legitimate question, raised by entrepreneurs, salespeople and digital marketers. Unfortunately, we often end up approaching the issue with the wrong approach . Those who try to give themselves an answer often do not have the right skills and tools to do so.
Google PageSpeed Insight is undoubtedly one of the most popular tools, praised by numerous Web Marketing articles and courses. Let’s take a step back, what is it? And above all: Is it really that important?
Google Page Experience Update, Core Web Vitals and the race for the best score
One of the major causes of the increased attention to Google PageSpeed Insights results in recent years is certainly due to the Google Page Experience Update of June 2021, in addition to the Mobile First Indexing of 2015.
With the 2021 algorithmic update, the usability of websites becomes a ranking factor and is defined with 3 main metrics, the fateful Core Web Vitals .
These are the now famous metrics that measure the loading speed of the largest element in the window (LCP), the time between the first user interaction and the moment the browser responds (FID) and the visual stability of the layout ( CLS).
It must be said that the FID will be replaced, in March 2024, by the new INP (Interaction to Next Paint) metric , which measures the overall responsiveness of a page to user interactions by observing the latency of all clicks, taps and keyboard typings. INP is currently in the experimental phase ( Pending ) and we can already find it on PageSpeed Insights.
It therefore seems that Google gives a lot of importance to Core Web Vitals, so let’s see how these integrate with a much coveted high score on PageSpeed Insights.
What is PageSpeed Insight and how does it work
According to the official definition of Big G, PageSpeed Insights (acronym PSI) is a tool developed to evaluate the performance of a page both on mobile and desktop, also providing practical indications for improving it.
Furthermore, PSI says it uses both laboratory and empirical data collected “in the field”. Laboratory ones are useful for performance debugging and are collected in a controlled environment.
While Google uses, when possible, empirical data (i.e. Field Data ) to evaluate Web Vitals , the score is assigned based on Lab Data , i.e. Lighthouse. It goes without saying that it is always advisable to evaluate the score with a critical and contextualized look .
Does it really make sense to plan hours-long interventions to achieve a high score, when in reality our users are not experiencing usability problems on our site? In fact, it is not uncommon to find scenarios in which you have low scores alongside a positive Web Vitals rating and vice versa . Similarly, we find sites in SERP with excellent positioning that have low scores on PageSpeed Insight and a barely sufficient Core Web Vitals rating.
Google itself also claims that the so-called Data Labs fail to capture bottlenecks that occur in the actual user experience.
This is why we must take the indications returned by the instrument with a pinch of salt. Let’s see other supporting reasons and also the results of a test we performed on some very famous sites.
The PSI score changes even without changing the site
John Mueller of Team Google reported that the PageSpeed Insights score can change without making any changes to the website you are testing with the tool.
This information was provided on Twitter in response to an SEO who questioned Mueller about the fluctuation in PageSpeed score. Mueller says these scores are not physical measurements and may change over time whether or not the website has changed .
Ultimately, Mueller believes these tools could be used to find ways and insights to improve a site in the eyes of users. Reaching a certain PageSpeed score, however, should in no way be seen as an end goal.
@JohnMu Thank you. From the various threads on this list in the last few days I see there indeed has been a bug + update to the PageSpeed Score. Was already wondering what happened as we did not change the website at that time.
— Sebastian Simon (@_sebastiansimon) February 5, 2019
This is a first explanation for the fluctuations in PageSpeed score that the Twitter user was alleging and referring to.
Pagespeed Insights vs WordPress
When trying to test a WordPress site, the most interesting statistic is the actual time it takes for the test to load the site. What assets were loaded, how long it took to load them, and what items were tested. Google Page Speed Insights, unfortunately, does not reveal all this information to you.
What does Google PageSpeed Insights actually measure?
Google PageSpeed Insights evaluates your site based on what Google believes are best practices, not necessarily the same best practices taken into account by WordPress and general web development rules.
If you take a look at the Google PageSpeed Insights test results page , you won’t see any data mentioning actual loading speeds – Google will instead analyze and rank your site based on its best practices.
It’s true: a high score is still somewhat correlated with a fast user experience but, moreover, without there being any guarantee of this.
We’re not saying that you shouldn’t aim for a high score in this tool or that their suggestions are totally wrong.
We are saying that you shouldn’t pay too much attention to the details of their test, try to focus more on your site’s actual speed and good user experience .
If users perceive your site to be fast and if you have content they are interested in: you will have won (even if it doesn’t look perfect on Google PageSpeed Insights!).
You may have already heard or read about some of these critical issues if you work in the tourism sector. PageSpeed Insights has often been named in this industry for generating negative buzz about poor performance . It’s part of Google’s effort to encourage high-performing sites on its tool . In particular, this has been accentuated with the increase in mobile web use since, apparently, PSI appears to be an easy way to compare performance with its competitors.
Trust the Experts
Problems and critical issues aside, PageSpeed Insights is a tool for developers. However, the tool has been used over and over again without this critical warning. It was not meant to be operated by people who don’t speak the technical language of performance. The Insights FAQ assumes that the average user is a webmaster.
The misuse of PageSpeed Insights is causing confusion . It’s making getting a 100/100 score from an online test more important than the actual experience of a real person visiting your website.
Improper use of this tool has real consequences for businesses. Generate discussions about reading the performance of the Internet as a whole. It is not educating and it is not providing solutions. It’s time to leave it in the hands of developers and use the metrics that matter to evaluate site performance.
So, unless you have a good understanding of Javascript and SEO, delete PageSpeed Insights from your favorites. Don’t waste your time chasing a grade that doesn’t matter. Focus on real SEO optimization factors and making the user experience better.
Do you know for sure that your website is too old, problematic and you need to redo it? Do you need professional SEO consultancy to be among the first search results?
WebQCoder is here for you. Do not hesitate! Tell us your projects , let’s make them grow together!
