Google PageSpeed Insights sounds like the perfect way to assess your website’s speed, right? After all, it’s free, it’s made by Google, and it has a simple scoring/color system. Wrong.
In reality, Google PageSpeed Insights is one of the worst ways to measure your site speed. Yes, that’s right. Your site’s Google PageSpeed score is largely inaccurate and useless.
Surprised? You might be even more surprised to learn that that fastest loading, top ranking sites on the internet have some of the worst Google PageSpeed scores.
For this reason, marketers and business owners should take the results with a grain of salt and take care not to fall into the trap of being obsessed with optimizing their site for this score.
Google PageSpeed Insights don’t reflect real world page speed
Google even says it themselves when you usually run a test. Most of the time, Google PageSpeed Insights have no correlation with how fast your site actually is. The colors and grades in Google PageSpeed Insights are not based on real-world speed testing so usually produce inconsistent and unreliable results.
It’s much better to look at your sight through the users eyes, or User Experience, and evaluate the site performance. Here’s an example.
A site has hero banner image at the top, which is the first thing the user sees. Below the fold (the initial screen area) the site loads the detailed assets, or uses a lazy load approach to load them as needed by the user as they scroll down the page.
In this scenario, it is important that the banner image is optimized for fast, lightweight loading, but other assets can load more slowly after the banner is loaded. In this scenario, the total page load time is largely irrelevant, as the user experience is not impacted by the slower loading assets below the fold i.e. out of their view. But you might be loosing sleep and time because PageSpeed Insights is telling you that your site is slow.
Tip: View your site through the lens of the user and their experience, not pure algorithmic speed numbers that don’t reflect reality.
PageSpeed Insights are based on automated checks
It goes without saying that there isn’t one size shoe that fits all feet. PageSpeed Insights don’t understand your specific website, your niche, your industry, or your goals. Instead, they rely on generic bits of code that return a simplistic yes/no. It often flags things that aren’t an issue for your specific setup, and exaggerates factors that have little to no effect on your real word website speed or, most importantly, user experience.
PageSpeed Insights were intended for something else
They were initially intended for highly technical web developers who needed to make final corrections to a website build. That’s why its suggestions are largely useless to the vast majority of sites. Despite its claims, the vast majority of the suggestions it makes for your site will have little to no discernible impact on your site speed.
It’s great for highlighting low hanging opportunities, like images that haven’t been optimized or slow pop-ups that hold up the rest of the page load, but that’s about it.
If you’re on Shopify, you’ll probably often seen recommendations made to optimize Shopify code and snippets, the world’s leading eCommerce platform focused entirely on delivering fast websites, which you have to agree, doesn’t really make sense.
PageSpeed Insights are inconsistent
One of Google’s senior analysts, John Mueller, noted that PageSpeed Insights can be inconsistent. He advised that the tool should be used to give an idea of places where improvements could be made for your users’ benefit, but the score shouldn’t be seen as a final goal.
If you’ve used PageSpeed Insights before, you’ll notice that running tests at different times and days can actually return different results, which may lead you to drawing incorrect conclusions about a great service or app or worse choosing a poor performing app.
PageSpeed Insights suggest experimental tools
As with many Google initiatives, PageSpeed Insights has a strong bias towards using other Google tools and programs. While that may not sound like a particularly bad thing at first, it’s worth noting that some of the suggestions are for experimental creations that may be discontinued in future or may not be compatible with all browsers.
PageSpeed Insights are bad for WordPress sites
While PageSpeed Insights are largely without benefit for most sites, it is particularly bad news for WordPress users. The most common reason for a website being slow is images not being optimized correctly. The second most common reason is actually something specific to WordPress, problems that won’t be identified by PageSpeed Insights. Some of the suggestions made by PageSpeed Insights can also be damaging to WordPress sites. For example, deferring particular JavaScript files, can really mess up your site.
PageSpeed Insights uses 3G (yes, really)
PageSpeed Insights uses a slow 3G connection for its mobile testing, rather than 4G or 5G or a regular broadband connection. This means that the large majority of your visitors will see faster page loads and have a much better experience browsing your website by default than PageSpeed Insights suggests.
Our conclusion? Don’t waste your time.
Let’s make something clear. Only the actual user experience is what matters.
Scores churned out by Google with red, orange or green colors really do have little or no bearing on your site speed. You could also waste valuable time trying to make recommended improvements, only for visitors and search engines to see no improvement to your speed.
Why not use the time you’ll save to implement a review collection strategy and make use of user-generated content across your store? They DO have a positive impact on traffic, they DO boost trust and credibility, and they DO have real potential to increase sales.
