Since it’s an automated test, there can be a gap between a page's actual speed, how fast customers think it is, and the performance score an algorithm gives.
For example, running the same site through tools like Pingdom Tools, GTMetrix, WebPageTest, and YSlow often gives different performance scores. We recently tested a product page that scored 77 and loaded in 2.2 seconds. But after adding apps and third-party services, the speed dropped to 3.0 seconds, and the performance score fell to 30, without reflecting the improved functionality.
It's important to remember that automated tools, like domain and trust scores in SEO, aren't exact sciences. They're guides to help prevent major issues and boost website speed incrementally across the web.
Also, some flagged areas might be out of even the best devs' control (ahem). Shopify apps and third-party services often inject code that we can't control, and Google Analytics can be a big culprit when it comes to site speed!