After evaluating hundreds of Shopify stores we have decided to stop using PageSpeed Scores to evaluate the speed of a shop.
There are a few reasons for this:
1. They are inconsistent
Since PageSpeed updated to v5 in Nov 2018, PageSpeed scores are very inconsistent.
We have found that a single shop’s score can vary by up to 20 points without changing anything in their shop.
You can read more about the PageSpeed v5 update here.
2. They are impacted by external factors
We discovered that most significant changes to PageSpeed scores over time were caused by external factors.
External factors cannot be controlled by a shop or an app.
3. They focus on compliance with best practices
It is very difficult to properly measure how fast a site will be.
Instead, PageSpeed scores predict how fast a shop is by evaluating how closely that a site’s code follows what Google has defined as best practices for speed.
As a result, a low score does not always mean that your site is slow. Focusing directly on load times or asset size will give you are more accurate picture of you shop’s speed.
You can read more about using other speed metrics here.
4. They are not Shopify specific
Google’s PageSpeed is concerned with the speed of every website globally. As a result, many of their recommendations are not applicable for a Shopify site.
For example, PageSpeed Insights often recommends that a site should leverage caching to speed up their site.
This makes sense for many websites but does not apply to Shopify sites. The infrastructure of Shopify does not allow for leveraging browser caching as noted here.
Do you have more questions?
You can email us at firstname.lastname@example.org.
We’d be happy to help answer any questions you have about using PageSpeed score and measuring speed.