Google Confirms Mobile First Indexing Program for All Sites

[engine]

Google has confirmed details of its mobile first indexing, and what webmasters should expect over the coming months as it reaches its goal of switching for all sites starting September 2020.

It said earlier that there should be notifications in GSC [webmasterworld.com], and now it's gone on to explain more.

When we switch a domain to mobile-first indexing, it will see an increase in Googlebot's crawling, while we update our index to your site's mobile version. Depending on the domain, this change can take some time. Afterwards, we'll still occasionally crawl with the traditional desktop Googlebot, but most crawling for Search will be done with our mobile smartphone user-agent. The exact user-agent name used will match the Chromium version used for rendering.

It's really worth reading the whole document to understand the changes we can expect, and where to watch out for notifications, such as GSC.

[webmasters.googleblog.com...]

[mosxu]

WOW

[Dimitri]

I wonder if all of us, who put lot of efforts since years to adapt /make our sites mobile friendly, from an UI/UX point of view, and the optimizations to server content fast, will see a difference.

[mcneely]

... while we update our index to your site's mobile version.

Wait! ... What? ... Did Google just make reference to "your sites *mobile* version"? .... Welcome to 2009 Google, we're just thrilled you could join us /sarcasm

Pretty sure that RWD (Responsive Web Design) made an extra *mobile version on your domain pretty much obsolete.

Of course then maybe Google was just making reference to Facebook, as it still runs with it's separate .m subdomain for all of it's mobile traffic

[samwest]

Of course then maybe Google was just making reference to Facebook, as it still runs with it's separate .m subdomain for all of it's mobile traffic

True enough, as many people now think their Facebook page IS "their own" website.

[RedBar]

I wonder if all of us ... will see a difference.

ROFL ... Just HOW would they KNOW since they can't detect scrapers nor the original article or image creators ... It'll be G's usual continuation of their selected garbage in, their preseferred and selected garbage out.

Or will I be proven wrong?

Whoops, just woke up out of dreamland:-)

[tangor]

Is there a difference between RWD and Mobile?

Really?

[EditorialGuy]

Pretty sure that RWD (Responsive Web Design) made an extra *mobile version on your domain pretty much obsolete.

Not necessarily, and what about AMP?

[mcneely]

... what about AMP?

AMP has turned out to be somewhat of a blow ...

[iamlost]

A lot of enterprise sites built m.site subdomains a decade ago and see no reason to change. Google is not about to rock their boat.

Once the ‘switch’ is complete it will be interesting to see how much truly changes from now. My suspicion is that not much , that mobile friendly, as site render speed, is more a differential tiebreaker when necessary not a major input.

[fearlessrick]

I'm just thrilled to be here.

[j_plenti]

I've been surprised that some of my sites have received messages that they are significantly different enough from desktop to mobile versions to still warrant desktop crawling. Guess I got some work to do...

[aristotle]

This whole matter has always been hard for me to understand. Why would desktop rankings be based on what mobile users see?

Anyway i converted my sites to responsive design years ago, and just checking GSC it says they are all mobile-friendly. But why does it say, for each page:
Status = Valid ... Validation =N/A

Does it mean that the date when its mobile-friendliness was checked is no longer available, or not applicable? Or does it mean that the method of validation isn't available or not applicable? I don't understand it.

[Geekytech]

Stilling LOLing at the mobile site comments. While I do believe that RWD has pretty much made the use of mobile sites absolutely obsolete, there are still a lot of companies that still opt to having the mobile varients instead of RWD usage.

Doesn't seem to make much sense to me, though.

[JorgeV]

Hello,

I guess that some companies see an interest in spending more to maintain two versions of their sites.

To me, it's more convenient to have a responsive site, which works on both wide and small screens. But my be for others it's not that obvious.

By the way, the advantage of a responsive sites is that , even desktop, if someone is resizing his browsers, it adapts the content.

[lucy24]

The exact user-agent name used will match the Chromium version used for rendering.

:: quick glance at most recent logs ::

Mozilla/5.0 (Linux; Android 6.0.1; Nexus 5X Build/MMB29P) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.96 Mobile Safari/537.36 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)

When you’re all finished ROFLing at the mobile-first thing, you can LOL all over again at ... Chrome/41.

[aristotle]

Maybe Chrome 41 was the version that was available when google first started using a separate crawler for mobile.
Probably they just haven't changed the old UA string yet to match what they're using now.

[lucy24]

Maybe Chrome 41 was the version that was available when google first started using a separate crawler for mobile.

Their earliest mobile crawlers--DoCoMo and Samsung, from around 2011--are no longer in use. For several years after that, they went through a series of iPhone UAs before belatedly (in 2016!) remembering the existence of Android. Chrome version 41 dates back to early 2015; by early 2016 Chrome was up to 48. (Information courtesy wikipedia.) Compare the various Googloid functions that formerly used Chrome 27 and have since moved to Chrome 41 ... and, so far, no further.

[iamlost]

I’m presuming that you performed an rDNS-fDND to confirm it truly is what the label says.

If so one must simply default to accepting that a tech company used to second fractional thresholds has found 27,316,000 seconds insufficient time to implement stated requirement.

This version number will increase over time to match the latest Chromium release version used by Googlebot.

Googlebot now runs the latest Chromium rendering engine (74 at the time of this post) when rendering pages for Search. Moving forward, Googlebot will regularly update its rendering engine...

[lucy24]

Oh, I don’t doubt that their rendering is up-to-the-minute. What gets me is what they put in the visible UA string--even at the risk of making it look so old, some sites would block them by default. But then, an awful lot of webmasters don’t seem to fully appreciate that the name a search-engine spider gives itself has almost nothing to do with what the search engine’s computers end up seeing. (It is “almost nothing” rather than “nothing at all” because some sites will serve different content depending on the UA of the visitor. But that only makes it worse, since site might say “I know Chrome 41 can’t do suchandsuch, so I’ll serve this other simplified old-fashioned content” ... and then the site is judged on the basis of that simplified old-fashioned content.)

I’m presuming that you performed an rDNS-fDND to confirm it truly is what the label says.

Was that directed at me? Robots that call themselves Googlebot but don’t come from their crawl range of 66.249.whatever-the-heck-it-is are simply blocked at the gate. (This has become less common in recent years.) Yup, that includes the Googloid range that lives immediately adjacent to the crawl range.

[iamlost]

@lucy24: the reason I always ask these days re- rDNS-fDNS is that Google no longer has dedicated crawler IPs and themselves recommend doing the check to be certain.

The old, as in this time last year, bot chromium version number would definitely have me doing a check if I already didn’t do it on every crawler on my whitelist automagically.

Trust but verify.

Actually, when it comes to web cooties I’m paranoid with extra tin foil.

[lucy24]

In the specific case of the mobile Googlebot it is always the identical UA.

[iniciopublic]

what is Mobile First Indexing?