When SEO professionals and webmasters have questions, particularly Google-related questions, some people like to ask Google for answers.
One such avenue for doing so is the Google SEO office hours, which are Q&A sessions where webmasters can ask questions to a Google representative (in this case, Search Advocate John Mueller).
During the most recent session, a participant described his site. Basically, he has a 1400-page site with product information and reviews.
He said that “…the first strange thing is that only the 20 percent of these pages are listed in the valid mobile usability list…even if the structure of the pages are more or less the same.”
So, I think that, by him stating that the structure of the pages is more or less the same, if one page, which has the same structure or template as most of the other pages, is of the 20% listed as valid in the Mobile Usability section of Search Console, then presumably, most of the other pages should be, as well.
Same Template for Most Pages, So Why Only 20%?
I’m guessing that this participant is reasoning that if all ~1400 of his pages use the same template/structure, then nearly 100% of his pages should be listed as valid in the Mobile Usability section.
This makes sense, and is sound reasoning; so why aren’t more of his pages marked as valid?
And Then, a 40% Drop
John Mueller eventually answered, but only after the participant described an intensification of this situation:
The participant went further, “…but even more strange, since 10 days ago, I saw a major drop about 40 percent in the number of the URLs that are listed as mobile valid–you know and without any site restructuring.”
The participant further went on to describe how he compared two lists of URLs (one from 10 days ago, and one from the day earlier), and basically found no explanation as to why there should have been a 40% drop.
John’s response was: “Yeah, I agree this is kind of confusing. What is happening there is, in Search Console, for these kinds of aggregate reports, we show a sample of the URLs from your website.”
That’s key. A sample, not all 1400 pages. (My guess is that this might save Google a lot of processing power. It may also save you some of the so-called ‘crawl budget.’)
What About the Sample Size?
But that leaves one lingering question: Why the drop of 40%? Doesn’t the sample size stay the same?
As part of his response, John said, “…you can think of it a little bit like you have 1000 URLs on your website, and we will double check 200 of them for mobile-friendliness, and that number 200 that can change over time.”
So, the sample size can change. This also applies to AMP and, in a way, Core Web Vitals reports.
“So it can happen that maybe one time we only check 100 URLs to see if they’re mobile friendly. That means the total mobile-friendly count [in Search Console’s Mobile Usability section] can go down, but that doesn’t mean that the number of pages that are mobile-friendly on your website go down.”
What You Really Need to Look For
John said, “In short what you need to look for there is more the number of errors and kind of the proportion of normal pages to errors that we report in Search Console.”
In short, just because you only see that a certain number of pages are marked as valid, doesn’t mean that those are the only pages on your site that are valid (in terms of mobile-friendliness), because Google is only looking at a sample (that changes in size over time).
Instead, what you really should be considering is the number of errors–or rather, the proportion of error pages to normal pages–that you see. Are those increasing, or decreasing over time?