Google Switches Mobile Search To Show Real-World Names

[engine]

More changes to mobile search prior to the mobile update coming on 21 April.

Now you'll see web site names instead of URLs in mobile SERPs.

To help mobile searchers understand your website better when we show it in the mobile search results, today we’re updating the algorithms that display URLs in the search results to better reflect the names of websites, using the real-world name of the site instead of the domain name, and the URL structure of the sites in a breadcrumbs-like format. Google Switches Mobile Search To Show Real World Names [googlewebmastercentral.blogspot.com]

Structured data site names and URLs
As part of this launch, we’re also introducing support for schema.org structured data for websites to signal to our algorithms:

The website name to be used instead of the domain name

The URL structure of the URL as breadcrumbs

Links to png images.
What it looked like before the change. [3.bp.blogspot.com...]

What it looks like after the change. [2.bp.blogspot.com...]

[goodroi]

Can't wait for the funny mistakes... Is it SpeedoFart.com or SpeedOfArt.com. I'm not even thinking of the NSFW mistakes.

Even without the funny mistakes, this seems to make it easier for non legitimate websites to spoof legitimate websites in the SERPs.

[webcentric]

Even without the funny mistakes, this seems to make it easier for non legitimate websites to spoof legitimate websites in the SERPs.

Was thinking the same thing! Imagine what the SERPs would look like if all that junk was shunned the way legitimate sites are today. Google is making it easier to spread malware and other garbage, not harder. Their ads are rife with it and this is going to make it easier for pretenders to hide in plain sight.

[piatkow]

Even without the funny mistakes, this seems to make it easier for non legitimate websites to spoof legitimate websites in the SERPs.

My immediate reaction too.

[aristotle]

Not all of the mistakes are funny, at least not to the owner of the site. Google SERPs sometimes changes a page title to something that misleads the searcher, mis-representing what the page is about, and resulting in fewer clicks and visitors.

[lucy24]

Now you'll see web site names instead of URLs in mobile SERPs.

Here's a really daring and innovative idea, google. How 'bout you have the SERPs show the actual title of the actual page that the actual designer actually used.

Naah. Too weird.

[goodroi]

@lucy24 You can't blame Google for changing titles. Too often publishers are big stupid idiots and publish content without changing titles or they use the wrong template which ends up mislabeling the page title. I have seen many US government web pages have an incorrect page title so it would be wise for Google to change that.

You and I know how to properly build a site so we don't like it when Google changes our hard work. I like it because it reminds me that Google isn't my friend. I should never drop my guard when dealing with Google which is ultimately a for-profit company looking out for their own interest and not mine.

[netmeg]

I have seen many US government web pages have an incorrect page title so it would be wise for Google to change that.

Too true. I wish I had a nickel for every city / county / state website whose home page has a title of... Home.

[ken_b]

I'm lost, does this script go in the head section of the home page?

Added found the answer.

[edited by: ken_b at 10:12 pm (utc) on Apr 20, 2015]

[EditorialGuy]

You can't blame Google for changing titles. Too often publishers are big stupid idiots and publish content without changing titles or they use the wrong template which ends up mislabeling the page title.

Or the titles are just lists of keywords that someone concocted after reading SEO for Dummies: 1999 Edition.

[aristotle]

In the vast majority of cases, there's no valid reason for Google to change the original author's title.

And even in the rare cases when there might be a reason to change it, Google's algorithm often messes it up and makes it worse.

[EditorialGuy]

In the vast majority of cases, there's no valid reason for Google to change the original author's title.

In the vast majority of cases, Google probably isn't changing the original author's title. (Or the SEO's title, as the case may be.)

Note that I said "probably." We have no way of knowing what's happening in the vast majority of cases, although Google probably does. :-)

[lucy24]

I wish I had a nickel for every city / county / state website whose home page has a title of... Home.

On my test site, the current page titles are
:: shuffling papers ::
a
new document
Blank page
new page.php
Untitled

But I did it on purpose.

[keyplyr]

Now that the mobile update is in force I'm seeing the names and breadcrumb changes. IMO it looks great, I really like it. Displaying the breadcrumb path gives the user a bigger picture and may offers a little insight into how the respective site hierarchy is built. The user has a little more understanding going into a new site.

[dannyboy]

Now that the mobile update is in force I'm seeing the names and breadcrumb changes.

I actually noticed this yesterday afternoon when doing a few searches. For many sites, the domain name (e.g., www.example.com) was still shown instead of the name.

This looks like it's been active for at least a week, if not a little longer.

More info here: [developers.google.com...]

[Wilburforce]

We have no way of knowing what's happening in the vast majority of cases

We can easily know in any individual case, and as long as a sample of cases is representative we can have a pretty good idea.

[Wilburforce]

More info here: [developers.google.com...]

Useful, thanks!

Noting that a name must "Be unique to your site—not used by some other site", I wonder what Google will do if more than one site claims the same name. That is bound to happen sooner or later.

I have named mine, anyway.

[lucy24]

wtf. Is that what this is all about-- not page titles but domain names? Would be nice if the referenced page [developers.google.com] said what they use if you don't happen to include their desired markup. Does it default to the real domain name or, like pages, do they make something up?

:: happily picturing a system in which any domain name that is more than 20 characters, or that contains two or more hyphens, shows up in SERPs as "spam site" ::

What happens when one site is example dot com, another is example dot org and a third is example dot net, and all three are generally known as "example" although serving entirely different purposes? *

I gotta say I don't care for the idea that every site should have yet another slab of structured data. Last time I looked, structured data markup won't validate-- meaning that you have to fine-tooth-comb to make sure there are no other issues-- and it's more bytes for the device to download.


* There used to be a pbs dot com. I believe they finally went out of business.

[Wilburforce]

structured data markup won't validate

I don't know about generally, but Google's script in dannyboy's link does.

I have put in on my homepage, and will wait to see what happens. However, I'm with you on the example.com/.org/.net question (although in my case my site name is a registered trade-mark, and I own domains with quite a lot of - but not all possible - extensions). I can see the potential for some intellectual property lawsuits in this one. If in doubt, get yours in early.

[Tonearm]

Should I have schema.org WebSite markup on every page or only on my home page?

This page says schema.org markup is not ready for breadcrumbs yet:

[developers.google.com...]

Is that correct?

[Wilburforce]

Should I have schema.org WebSite markup on every page or only on my home page?

In this particular case, only on your home page. In general, anywhere and everywhere as you think fit.

Arguably, in some instances it is likely to influence ranking factors like CTR, and broadly should help search-engines to figure out what a page or site is about, but I haven't any evidence of clear beneficial effect. It is extra code with no presentational effect.

[samwest]

One more way for GORG to remove our individuality and uniqueness. I can just see my competitor who already mimics my my site name and page title stealing more of my identity. This is tantamount to shaving our heads and tattooing us with their mark. They've already taken all of our personal effects and gold teeth fillings. Switch to BING before it's too late.

[Wilburforce]

One more way for GORG to remove our individuality and uniqueness.

The effect of this should be the complete opposite. How would a name that must "Be unique to your site—not used by some other site" make a site less unique?

What your competitor is doing has nothing to do with Google, and by the sound of it you need to do something about that yourself.

[lucy24]

:: detour to google and schema.org, which has the usual effect of leaving me angry and frustrated* ::

<div itemscope itemtype = "http://schema.org/WebSite">
 <meta itemprop = "url" content = "http://www.example.com/">
 <meta itemprop = "name" content = "Example">
 <meta itemprop = "alternateName" content = "the real Example and all the others are just spammers and fakes">

What's a meta doing inside a div? Where does the package go?

Incidentally I now realize I was misunderstood up above. When I said "won't validate" I didn't mean google's Structured Data tool. I meant the w3 validator. If you feed it something like

<div itemscope itemtype = "http://schema.org/WebSite">
 <meta itemprop = "url" content = "http://www.example.com/">
 <meta itemprop = "name" content = "Example">
</div>

as an HTML fragment, it screams loudly (three times for un-recognized elements, twice for a meta inside a div).


* Exactly the same feeling as when trying to extract information from lilypond manual: Yes, OK, I understand what I'm supposed to say ... but where the ### do I say it?

[aristotle]

Lucy that Schema.org code is clumsy and torturous. Whoever created it must be a total nincompoop. It's ridiculous that you would need that much code to specify a site name.

[lucy24]

The irritating part is that they already know your site name. I was doing some random searches earlier today (not checking anything, just real life) and happened to notice that the first line of each search result already gives the site name. Sometimes you can't tell, because the pagename alone is too long and then the rest gets chopped off. And sometimes apparently they couldn't figure it out. But if everything is concise, it goes pagename, hyphen, sitename. Dunno what the underlying code is, but in the one result that happened to belong to my site-- and in some of the others I randomly checked-- they simply used the <title> of the front page. For any normal site, that would be the site name-- and the ### thing's already there in the html.

[charlesdavis]

I think it is good decision taken by Google because if your website is responsive so ultimately you get more searches from users.As we know the 80% people of World having mobile phone and it is very approachable device to see instantly anything.

[Wilburforce]

@lucy24

The microdata version you have used won't validate (and Google's own suggested version of it doesn't). If there is a way to state the site name using RDFa - which will obviously validate - I don't know what it is.

However, inserting Google's JSON-LD script in the HEAD section of your homepage (probably Google's preferred method, as it comes first) passes Googles test and validates on W3C (HTML5):

<script type="application/ld+json">
{ "@context" : "http://schema.org",
"@type" : "WebSite",
"name" : "Your WebSite Name",
"alternateName" : "An alternative name for your WebSite",
"url" : "http://www.your-site.com"
}
</script>