Discerning the Real Value of Google PageSpeed Insights Test Results

 


Sep 08 02:49 pm
Written by: Pete Bruhn

 

Testing Website Load Speed

 

We all know from experience how important website speeds are, especially when using a mobile device over a mobile network. As technology advances, the importance of delivering a convenient and streamlined experience, facilitated through reliable and quick-loading webpages, is more crucial than ever.

 

There are tools out there, like Google PageSpeed Insights, which can test and grade the performance of your website. While they can be helpful, they don’t always paint a true picture of how a website performs, either from a load speed perspective, or from a user experience point of view.

 

Because there’s been a recent influx of questions related to site speeds, particularly as they relate to Google’s PageSpeed Insights tool, let’s spend a few minutes discussing what these tests mean and how to speak to their results.

 

Breaking Down a Google PageSpeed Insights Score

A website that scores a 100/100 on mobile or desktop speeds using the test on Google PageSpeed Insights doesn’t necessarily mean that the user experience is amazing or that the website performs any better than a site that gets, say, a 75/100. In fact – and this might come as a surprise to some of you – Google PageSpeed Insights scores aren’t truly accurate because they don’t, and can’t, grade a user’s true site experience. Let’s elaborate.

 

First of all, PageSpeed Insights results are just a subjective score based on general criteria that Google recommends to the average website/webmaster. It’s not the authority of website speed analysis. While it’s a great reference guide, it doesn’t account for user experience as it relates to a website’s loading time. Some things that are slow in automated tests may not be impacting actual user experience (like background events that the user never sees).

 

Site speed tests such as Google PageSpeed Insights do not account for server-side processing, and instead just examine network latency. In short, they don’t analyze the true experience a user has while loading websites. In addition, the grading systems used by these sites don’t consider optimizations that improve the total download time at the expense of the first byte time, nor do they include the amount of offline caching that occurs, which increases user experience and speed upon page load.

 

Automated performance evaluations are general tools for use across a broad array of websites. As such, they focus on easily measurable, quantifiable metrics. While the outcomes of these tests are important to address, they’re probably not the best data points to use to inform a path toward website optimization.

 

Getting an Accurate Site Speed Score

If a Dealer.com website is loading slowly, the majority of the time it’s due to third-party content/integrations, custom CSS, and/or JavaScript. When using Google PageSpeed Insights, test the domain using “?_mode=reset” as a parameter on the URL. Adding “?_mode=reset” to the domain strips out all custom CSS, JavaScript, and third-party content. The typical results show a significant increase when testing with this method.

 

Below is a real example of a Dealer.com dealer website that shows the increase in performance when the third-party content/tools and the custom CSS and JS is stripped out:

 

Results without ?_mode=reset:

Mobile – 58/100

Desktop – 75/100

 

Dealership website with a poor speed score

 

 

Results with ?_mode=reset:

Mobile – 70/100

Desktop – 84/100

 

Dealership website with an okay speed score

 

Poor page speed scores aren’t just a problem for auto dealer websites, however. Take a look at how Google PageSpeed Insights scores two of the world’s most popular websites:

 

Amazon's poor speed score

 

eBay's poor speed score

 

 

 

 

As is usually the case with a variety of Google tools that test a website’s digital performance, this site isn’t considering all of the factors that go into an optimal user experience. As such, Google PageSpeed Insights tests are causing misperception across industries. But, what Google PageSpeed Insights do accomplish is to shed light on the importance of weighing the value of third-party content/tools and custom CSS/JS. There is always a trade-off between incorporating those features and the speed of your website.

 

Have questions or concerns about site speeds and loading times? Connect with me by using the comments field below.

 

Pete Bruhn is the product manager – website platform at Dealer.com

 

 

Leave a Comment