Site Speed

Google’s Core Web Vitals (CWV) metrics are typically looked at through a sample of traffic to the overall website. Therefore, it’s the pages on your site that get the most visits that will contribute the most to the overall CWV score. Having pages on your site that perform poorly for Core Web Vitals, but don’t bring in a lot of traffic, is going to be less of a concern. In the same vein, pages with little traffic and great CWV metrics aren’t likely to pull up the site-wide score. The exception is if Google has enough data to segment a certain part of the site and treat it separately. For example, a super fast blog with lots of visits may end up being looked at on its own, away from the rest of the content on the site.

As page experience rolls out as a ranking factor on desktop, John is keen to clarify that the weight it holds will largely mirror what’s seen on mobile. If it’s clear that a page is the best result for that query, page experience signals could be downplayed. If there are multiple pages in SERPs that could answer the user’s query and intent equally well, page experience is one of the factors that could be used to distinguish between them and rank one site above another.

John confirmed that as AMP is not a ranking factor, removing AMP pages shouldn’t have a direct impact on SEO. However, users should consider the potential knock-on effects caused by a drop in page speed (AMP pages tend to be faster than non-AMP ones, but that’s not a guarantee. There’s also nothing to suggest you can’t make regular pages very fast). AMP pages also tend to be crawled a little heavier than non-AMP, so you might also see a rise in crawl activity across the rest of the site.

John warns users not to expect significant rises in visibility from speed improvements alone. This is true even if the speed goes from very slow to very fast in a short space of time. Given the complexity of Google’s algorithms, changes that affect page speed in isolation are likely to have a much more subtle impact on visibility.

Google’s Core Web Vitals look at data that is delayed by around 28 days. This means that any significant page speed improvements you make on your website will typically take about a month to show up in the search results.  

John is keen to clarify that both technical SEO and content quality are extremely important. When it comes to the quality of content, Google is looking at more than just the text on a page.  Additional factors ranging from page design to speed can also have an impact on how Google determines the quality of your site.

Neither the type of website nor industry vertical will alter how much weight is given to Google’s Core Web Vitals metrics.

The page speed metrics Google uses to rank pages is based on a real user’s rendered experience, not simply the fetch time of a page.

Page speed mostly affects websites in search which are very slow. You probably wouldn’t see a ranking difference between sites which are reasonably fast, but you might if there is a noticable difference in speed.

When displaying AMP pages, Google is able to pre-render and cache them directly from the search results, which saves the time typically spent retrieving the initial HTML to get the rendered version. However, this isn’t something that is taken into account in the testing tools.