Google Search Console: False positive "mobility issues" frequent

[djbaxter]

I had this discussion on the Xenforo support forum and just updated it today.

You can safely ignore these "warnings". See Google Search Console reporting less mobile friendly pages

Case in point:

I just received this email from Google tonight (Oct 20, 2019, timestamp 7.07 pm EDT) regarding the contact us page of one of the sites I manage:

The following issues were found on your site:

Clickable elements too close together

Text too small to read

That site is not new. It's a responsive WordPress site for a small business and it has seen a reasonably good amount of traffic for a local rural small business. No one has ever complained about usability issues of any kind. The only thing that might remotely qualify are the AddThis social share buttons at the bottom of pages or posts. Remember this is the Contact Us page. There's not much on that page except a contact form and the sidebar.

But here's the clincher:

Immediately after receiving Google's email, I went to the Google Mobile Friendly Test site and entered the URL for that page. Here's what Google's own Mobile Friendly Test says:

Tested on: Oct 20, 2019 at 7:56 PM (49 minutes later, during which absolutely nothing about the site has changed)​

​

Page is mobile friendly​

This page is easy to use on a mobile device ​

​

Google's own mobile test site disagrees with the Google Search Console algorithm.

So which are you supposed to believe? An automatic algorithmically-derived email from Search Console? Or the evidence from Google's own test site and your own eyes and those of the site owners and users?

I always double check when I get one of those emails but I don't believe them. Note that in Search Console it identifies the source of the purported "errors" as the "Smartphone crawler" (at least if you've been moved to the Mobile First index which most by now have).

Try this: after you get one of those emails, go to the Search Console and just click on Validate Fix (without actually changing anything). I've just done this for the latest email but when I've done that for other sites I'll get an email back in a couple of days saying thanks - it's fine now. I'll let you know what happens in this case.

UPDATE
As noted above, after receiving that email, i changed nothing on the site. I simply went to the Search Console, clicked on the link to say I had fixed it for both listed "errors", and waited for validation of the fix.

I checked Search Console again today. Remember that I changed absolutely nothing on the site. Today Search Console indicates that it had rechecked and validated my "fixes" and now it confirms that the mobility issue no longer exists:

 

[djbaxter]

Also got the emails tonight:

Mobile Usability issues successfully fixed for site XXXXXXXX

To owner of XXXXXXXX

Google has validated your fix of Mobile Usability issues on site XXXXXXXX. The specific issue validated was 'Clickable elements too close together'.

1 pages on your site were validated as fixed.

and

Mobile Usability issues successfully fixed for site XXXXXXXX

To owner of XXXXXXXX,

Google has validated your fix of Mobile Usability issues on site XXXXXXXX. The specific issue validated was 'Text too small to read'.

1 pages on your site were validated as fixed.

 

[Rich Owings]

One possible cause of this is that a supporting file (CSS, etc.) is temporarily unavailable, triggering the warning.

 

[djbaxter]

One possible cause of this is that a supporting file (CSS, etc.) is temporarily unavailable, triggering the warning.

But I think that would trigger a different warning. Or it should. This is not an isolated occurrence. Every case of "mobility issues" I have seen has been like this. I just better documented this latest one to make a point that the warnings are usually bogus - and that their system is obviously seriously flawed.

 

[Rich Owings]

Yes, they are probably 99% false positives, especially in the last year or two. This might shed some light on it:

 

[djbaxter]

Thanks, @Rich Owings

That actually confirms that CSS not loading would create a different error than the ones I addressed in the OP. But it doesn't actually address the issue of false positives.

 

[Rich Owings]

Hmm, I recall seeing John Mueller commenting on this during a Webmaster Hangout, but can't find it.

 

[Wirenut]

This has been happening to me for a long time, exactly as you explained it. I always just figured it was a script or something that didn't load properly.

 

[Rich Owings]

One followup here... It's important to always check. I received a notice this morning and it was a false positive (as usual) but it was for a dummy testimonial page the dev had left live, complete with plastic surgeon testimonials on a dentist's site!

 

[djbaxter]

I agree. I do always double check.

 

[Rich Owings]

A new statement from John Muller on this issue:

Google determines mobile usability results based on the ability to render pages in a way that matches what a user would see on their device. Sometimes they may face issues when fetching the CSS or JavaScript files and this will display a small number of issues, with regards to mobile usability, in Google Search Console. However, John advised that these are usually based on temporary fluctuations in Google's ability to fetch individual files, but it will not affect the indexing of these pages.

Google Webmaster Hangout Notes: November 26th 2019 - DeepCrawl

 

[djbaxter]

Sometimes they may face issues when fetching the CSS or JavaScript files and this will display a small number of issues, with regards to mobile usability, in Google Search Console.

In my experience, it's much more than a "small number of issues" but it's nice to see it acknowledged.

 

[Wirenut]

Funny you bump this thread now, yesterday I got an email saying that I had 2 issues with a page on my website. I clicked on Validate last night without making any changes. This morning I see that it passes again. The same thing happens every few weeks.

 

[Tim Colling]

It's a little bit like the old fairy tale about "little boy who cried wolf", isn't it?