English Google Webmaster Central mobile-sites office-hours hangout

JOHN MUELLER: All right. Welcome, everyone, to today’s
Google Webmaster Central office hours Hangouts. My name is John Mueller. I’m a webmaster trends analysts
at Google in Switzerland, and I’m trying to bring across
some of the information we have around mobile sites at the
moment and hope some of this will be useful for you guys. I have a short presentation
planned, but if any of you want to ask a first question
around what mobile sites, feel free to jump
on in and go ahead. AUDIENCE: I have
a question, John. JOHN MUELLER: Okay. AUDIENCE: Will the
HTTPS certification help mobile sites also
rank slightly better? JOHN MUELLER: Yes. That’s essentially
independent of anything from mobile or desktop. AUDIENCE: Cool. AUDIENCE: But does the
mobile ranking factor only impacts mobile search results
or smartphone search results? JOHN MUELLER: I’m not exactly
sure what you mean there, Barry. AUDIENCE: So there is a
mobile ranking factor, right? Like a smartphone
ranking factor? JOHN MUELLER: I
mean, primarily what we do is demote sites
that we can tell don’t work well
on mobile at all. So that’s– AUDIENCE: You
don’t demote sites? [INTERPOSING VOICES] JOHN MUELLER: Mobile
ranking factor, yes. AUDIENCE: You don’t
demote the sites that are desktop that are not– JOHN MUELLER: Exactly. Yeah I mean, this is something
that only makes sense to do for those users
that are affected. It wouldn’t make sense to
demote a site in desktop search if it’s just bad in mobile
search or for mobile users. AUDIENCE: Thanks. JOHN MUELLER: All right. Let’s see if I can make
this presentation work. I can find the right tab and
share something interesting. So some of you, I guess, will
have seen some of this before. For others, it might
be kind of useful. I tweaked it a little bit
to include some of the newer things that we’ve
been seeing recently. So let’s go ahead
and get started here. In general, what we
recommend is that you also use a smartphone to access your
website so that you see what users would actually see when
they access your website. That’s not particularly
new, but it’s something that I think makes
a lot of sense, because people
tend not to notice how bad their own site is on
mobile because they rarely view their own site
on a smartphone. But lots of people
are using smartphones, so definitely go ahead and
get started on that as soon as you can. We have a lot of
recommendations, put together a lot of
technical information on building smartphone
friendly websites. I recommend going through
that as much as you can, as much as it
makes sense to you, so that you at least
understand a little bit more about the basics of
what we’re trying to do and the basics of our
recommendation about the things we recommend to avoid as well. With mobile, one
thing to keep in mind is we’re primarily looking
at smartphones at the moment. Feature phones
are something that tend not to be as relevant
anymore for mobile search. There are definitely areas
where feature phones are still very prominent, but
for the most part, we’ve noticed a lot of people
try to use smartphones, and that’s what they have
the bad results, essentially. Tablets are less of a problem,
because usually they can just view the desktop site just fine. We have three configurations
that we support. For the most part,
you can just pick whichever one
works best for you. We have one that we recommend,
but essentially all three of these can rank just as well. So it’s something where we try
to bring a recommendation out so that people who don’t
really know which one to go for have some guidance there. But essentially you
can pick whichever one works best for your website,
for your infrastructure, for your web master, whatever
your considerations are. So essentially we differentiate
between two variants. On the one hand, you use the
same URLs for your mobile site, or you could use different URLs. So different addresses for
the mobile friendly site. Sometimes for
different addresses they’ll do things
like an “m dot,” where you have your main
domain, main website on dot-dot-dot, for example, and your mobile version
on to show that this is for mobile. That’s fine. Just use whichever
variation that you want. If you’re using
the same URL, you can choose to either serve
the same HTML content or to serve different
HTML content. Same HTML content we call
responsive web design, where essentially
your website works on a variety of devices
and screen sizes. A dynamic serving is when
your website dynamically changes the content that
it sends to the user, depending on the device And
for different mobile URLs, you can use either one. Most commonly, you just
use different HTML there. So it will be similar
to dynamic serving except on different URLs. AUDIENCE: John, we’re
having an argument about this slide in the chat. You mind if I ask a question
while you’re on that slide? JOHN MUELLER: Go for it. AUDIENCE: So there’s
an argument saying– some say that Google says
responsive will rank better than the other two formats. Is that true or not? JOHN MUELLER: That’s not true. AUDIENCE: Okay, thank you. JOHN MUELLER: So you can
use any of these variations. We recommend
responsive web design primarily for technical
reasons, because there are fewer things
that can go wrong. But essentially you can
use either of the three, so whatever works best for you. Sometimes there is a CMS
behind your website that makes it easier to use
one or the other of these. Sometimes that makes sense to
kind of take that into account. But essentially
that’s up to you. And ranking-wise, they can
all rank just the same. Webmaster Tools is really
helpful for mobile sites, because there’s some additional
information in Webmaster Tools that you might miss othewise. So I’d recommend
using it in general, but I definitely recommend
using it for mobiles sitres so that you have all of
the information you need. Specifically around
smartphone crawl errors, if there’s something specific
to your mobile site that’s not working right,
Webmaster Tools can generally help
you a little bit. PageSpeed Insights is
another really useful tool that we have available. It gives you a
screenshot of your site as it might appear on
a smartphone and scores your site on the one hand
based on speed metrics– so what you could be doing
to improve the speed– and also to improve the user
experience on a mobile site. AUDIENCE: I’m
obsessed with that. I’m obsessed with that. I just wanted to
ask you a question, though just in regards to
the– I see some of my clients have above 83, but does
that help to have above 83? I mean, does that
help ranking as well? JOHN MUELLER: At
the moment, that’s not something we take
into account directly. So there are things like around
the user experience where you have indirect
effects, where you might see some changes where if
users are using it on mobile, they can recommend
it to other people. One element that’s
involved there, which does have an effect
in the search results, is whether or not your site uses
a lot of Flash, for example. I believe that also shows
up in Pagespeed Insights. And if your site
has a lot of Flash, we might show the little Flash
badge in the search results or say that this is primarily
a Flash site and kind of make it clear to users who
are using smartphones that this site probably isn’t
going to be as useful for them because they can’t really use
it if they don’t support Flash. AUDIENCE: Google already
did say that it’s better to have above 83, right? JOHN MUELLER: I don’t know
what the threshold there is. I think that’s
something you’d have to look at yourself
with those sites, especially around
user experience. It’s really hard to say
this number is good, this number is bad. But it’s something
where I try to look more into comparative metrics, where
you find a site that you really like, that you think
works great on mobile. Check that score out, and say,
I want to aim for at least that. Or as you’re working
on a site, show that you’re improving
on that score. So you start off with,
like in this case, 46, and you came out with
71 or 85 or whatever. So it’s showing the improvements
that you’re doing there. Like I mentioned,
Pagespeed Insights also takes into account user
experience elements, so things like whether or not
the page is of [INAUDIBLE] so that it works well on
smartphones, how big the user interface elements
are on that page. And those are things that
users will generally notice, but which also make it a
lot easier to recognize that the site is doing things
properly for mobile users. It doesn’t have two pixel
titles or links to click on. It actually has buttons that
users with normal sized fingers can tap on. One thing we haven’t
mentioned so much in the past, but I think is picking
up a little bit in the sense that we’re
noticing that this is a bigger problem than we initially
thought, is robots have text issues
around mobile sites. So this isn’t
something that users would see, because users
don’t look at the robots.txt before they access a page,
but we see that a lot when we crawl these pages. Essentially so that
we can recognize that a site is
mobile friendly, we need to be able to
crawl it in a way that we could see the
mobile friendly page. So that includes things like
being able to crawl the CSS, being able to crawl
the JavaScript, especially if the
JavaScript is used to create a redirect to
your mobile friendly pages. So if CSS, JavaScript, or
the mobile friendly pages are relevant, then that’s
something we’d need to crawl. Sometimes people also block the
crawling of their [INAUDIBLE] pages, for example,
because they say, oh, Google should index
my main desktop site, and again, that’s
a problem for us. We see a redirect to a page that
essentially we can’t look at, so we can’t tell if this is
really a mobile friendly page, and we can’t really take that
into account in the search results. So as much as possible, if you
have a mobile friendly site, I definitely recommend letting
us crawl CSS, JavaScript, and of course the
mobile friendly pages. Interstitials is a common
problem that we’ve seen, where we try to crawl
page with a smartphone, and instead of
getting content, we get an interstitial
that says, hey, you should download our app instead. And from our point
of view, what happens there is we find
this interstitial. We index the interstitial,
maybe on a separate URL, and we can’t
actually find a connection to your desktop pages. So we are disconnected
from recognizing the connection between
the desktop pages and the mobile pages,
and we can’t rank that appropriately in search, then. We can’t see that this is
actually mobile friendly page. Broken redirects is something
we’ve sometimes seen, where if you have different URLs
for the mobile friendly page and for your desktop page, and
you redirect from one version to the other. If you redirect
to an error page, that’s a really bad
user experience. Sometimes we see soft
404 pages as well, where you access a page
in the search results, and it brings up an
interstitial that says, hey, this page isn’t
available for mobile. And in a case like
that, the best solution is really just to keep the
user on the desktop page. If you don’t have a
mobile page for that, let the user at least
see the desktop page. So that’s something
to keep in mind. Faulty redirects is also
a really common problem that we’ve seen. We’ve sent out lots of
messages to webmasters about this, where
essentially someone clicks on a desktop URL in
the search results, and instead of seeing
that specific page in the mobile version,
they get redirected to the home page of
the mobile version. So you click through to some
article, on, I don’t know, SEO Roundtable, and
essentially, instead of getting that
specific article, you get redirected to
Barry’s homepage, which is a really bad user experience. And it’s almost
impossible for us to recognize the connection
between the desktop and the mobile pages, so
that’s one reason we send out a lot of messages about this. Unplayable videos, Flash
content, obviously, as we mentioned before,
if this is something that doesn’t work on mobile, try
not to show it to mobile users. For videos, there
are alternatives, like HTML5 compatible videos,
or HTML5 video players that download the right files
for the right devices. So that’s something
where it makes sense not to serve the Flash
content to mobile users. If you prefer to use an app
instead of a mobile website, that’s also possible
now, in the sense that you can connect pages
from your website to your app. And depending on how you do that
with this app indexing set up with the site maps and
the different– what are they called– action links
between the app and the page on your website, we can
actually show a link to your app in
the search results so that users can go
directly to the app or they can download
the app instead of going to your webpages. So that’s something probably
worth keeping in mind if you want to create an app. If you don’t have
an app, obviously maybe look at this in the future
when you do create an app. But it’s worth keeping in mind. One thing that’s fairly
new is in Chrome. It’s really neat
in how you can look at pages now with
a mobile device. So to emulate different types
of smartphones, different sizes of tablets or
different phone types. And you can also emulate
different network situations, so if a user’s on
Wi-Fi or using 3G, and try your site out like that. So that makes it a lot
easier to double check how you set your site up to
see that it actually works. And if you’re looking
at different themes, for example, for your
blog or for your website and you have a theme directory
that you’re going through, this is something that makes it
a lot easier for you to double check that these themes actually
work fairly well on mobile. All right. So kind of through
the list here. Like I mentioned, there
are different ways you could make your
mobile site work. All of them are possible
from our point of view. From a ranking point
of view, they’re all essentially equivalent. We recommend using Webmaster
Tools and Pagespeed Insights so that you get all
of the information you need from our side. Avoid blocking crawling of CSS,
JavaScript, or mobile pages. If you have separate URLs,
avoid app interstitials. Make sure you redirect
correctly, and avoid technologies that
don’t work on mobile. So if you have a
Flash site, maybe it’s worth setting up
either dynamic serving to serve an HTML5
version of the site, or maybe even moving to
something newer overall. And of course, try your
own website out yourself, or at least user
a Chrome emulator so that you’re sure that
it’s more or less works. And that’s essentially
it for my side. [APPLAUSE] Do you guys have any questions? AUDIENCE: Yeah,
faulty redirects. So is there a limit,
I mean, if a site has, let’s say, 20 faulty
redirects on an ongoing basis, is that going to eventually–
because Pierre did touch on that. He said that it can
affect ranking, right? JOHN MUELLER: Yeah,
so I think pages that have a faulty
redirect, I believe we demote at the moment. We show this label, “may
redirect to the home page,” and you can click
on “try anyway.” So for individual
pages, you’ve probably seen a lower visibility of
those pages on smartphones. For the site on a
whole, I don’t think you’d see any big change there,
because we know that this is an issue that is primarily
an issue on individual pages, not across the whole site. So it makes sense
to try to treat that as granular as possible. AUDIENCE: So once it’s fixed,
it’ll go back to itself, right? Or is it demoted for– JOHN MUELLER: Yeah. As we re-crawl and
re-index those pages, when we can see that it’s
redirecting correctly, we’ll remove that label
in the search results, and it’ll rank normally again. AUDIENCE: Okay. Thank you, John. AUDIENCE: Hey, John? JOHN MUELLER: Sure. AUDIENCE: So Barry and I were
debating this responsive design thing, and I’m confused
at your answer a bit. I mean, I understand
that Google will allow you to use each
three methods, but back on November 18, Pierre Far
and you were in a Hangout, and Pierre Far said that
responsive design had both, and I quote, “second order
wins and first order wins.” And on June 27 of this year,
you mentioned, and I quote, that “the responsive gives
a boost,” and I quote you, “you kind of credit your
website with that extra value you are providing,” end quote. So it still seems to me based on
that, that not only does Google prefer responsive, but it might
even give you a slight boost, or there’s some kind of SEO win. Or maybe not messing up the
redirects or– I don’t know what it is, but– JOHN MUELLER: It’s definitely
not something where we’d say if we recognize
the site is using responsive design, or if it’s
using some other method, we’ll promote the
responsive version higher. Essentially we treat
them equivalently. There are just some aspects
of responsive web design– for example, by using
the same URL– that make it a lot easier
to avoid mistakes. So you don’t have this
issue with the redirects, you don’t have the multiple
URLs, that could potentially be linked to where you’d
have to watch out for the rel canonical pointing
back to your main URL. All of those issues
kind of fall away. So there is a lot less,
I’d say, potential problems that you can cause yourself
by having one single URL. And of course, you
can check this page a lot easier on desktops just
by choosing a smaller view port, for example. It’s not something where you
have to actually try it out on a mobile device to make sure
that it’s working correctly. So I think there are a
lot of advantages there, but it’s not that
we’d say there’s an inherent ranking advantage
by using responsive web design. AUDIENCE: Cool,
thanks for the answer. Barry is now very
happy that I’m wrong, and he can bug me about it. JOHN MUELLER: Good. Well, I mean, I think it’s
not a matter of who’s right or who’s wrong,
but it’s important that you guys have all
of this information so that you can take
that into account when working with websites who
want to move to mobile. AUDIENCE: Oh no, for Barry,
it is a matter of who’s right and who’s wrong,
especially when it’s me. AUDIENCE: It’s OK to
be wrong sometimes. JOHN MUELLER: I mean,
everyone is wrong sometimes. And these things can
change over time. So it’s good to keep asking
us these questions just to make sure that you’re
not missing anything subtle there has
slightly changed. AUDIENCE: Speaking
of wrong, Matt Cutts just himself switched
from HTTP to HTTPS. He used a 302 redirect. Is that the wrong approach? JOHN MUELLER: I don’t know. I don’t know what
Matt is doing there, so it’s possible that he’s
still looking into this or that he’s just
trying something out. 302 could work there as well. It kind of depends on
what you’re trying to do. For the long run, we
definitely recommend using a 301 so that it’s
clear that you really want this other
URL to be indexed. But if you’re trying
things out, 302 is a good way to
do that as well. In the long run, I
imagine our algorithms will try to figure out what
this 302 actually means. So if you were to
leave it at 302 and it were to stay like
that for, I don’t know, a year or whatever, then
at some point we think, well, this isn’t really
a temporary redirect. It’s permanently there. And we might treat
it as a 301 anyway. But there are lots
of possibilities, and I can’t read Matt’s mind. So maybe he’s just
trying things out. And it’s good to see more
sites trying try to move HTTPS. I think that’s not
always trivial, and there’s some things
involved there that people just have to try out and
take step by step. AUDIENCE: John, I’m
just on secure browsing. You hear me? JOHN MUELLER: Sure, yeah. AUDIENCE: Secure
browsing is obviously very popular at the moment. Is that basically the–
it’s not any kind of boost or a benefit from
having a secure version, but from forcing a
secure version, rather? So a 301 you get rather
than just saying, well, I do have an HTTPS which
you can go on the site. We’re using it all the time. JOHN MUELLER: We’re
essentially looking at the page as it’s indexed. So if you happen to also
have an HTTPS version, but that’s not the one
that’s being indexed, then we wouldn’t take
that into account. AUDIENCE: All right. So it needs to
replace it, basically. JOHN MUELLER: Yeah, exactly. All right. Let’s go through the
questions and see if we have any mobile
ones there to keep everything themed together. Is there a preferred
method of mobile site currently can serve up
totally different versions of the website or have
the website responsive? Which is preferred by Google? We look at this previously,
and essentially, those three options are up to you. Whichever one
works best for you. And go with that. So it’s not something
where we say there is a ranking advantage. We recommend using
responsive web design because there are just fewer
things that can go wrong. It’s a little bit easier to
keep up with technically. But essentially, you can
use any of these methods. AUDIENCE: John, isn’t
it, technically, isn’t it heavier to
have a responsive site? JOHN MUELLER: It depends
on how you set it up. I mean, it sends the
full HTML, but it kind of depends on how you set it
up, how you use images. It doesn’t necessarily
have to be a larger site. But of course, depending on
the type of your website, there might be a lot involved
in actually keeping something really sleek and responsive. So it depends a little bit on
your website and a little bit on what you actually want to
provide on the mobile version or what you don’t
want to provide. So if, for example,
your desktop page has a lot of banners
on it, a lot of ads, a lot of images in the
sidebar, those kinds of things, a lot of extra HTML in
it that you don’t really need to serve on
mobile, then maybe it makes sense to strip
that out and go with something dynamic serving
or with separate URLs even to really have a
sleek mobile version. But if you have a really sleek
desktop version already, then maybe there’s not much that you
need to strip out for mobile. It kind of depends
on your website. AUDIENCE: Cool, thanks. AUDIENCE: I have
an HTTPS question. JOHN MUELLER: All right. Go for it. AUDIENCE: If it’s
[INAUDIBLE] on the questions. About a week ago, Searchmetrics
came out with a study reporting that in
their analysis, there is no relationship at
all between any ranking boost on the HTTPS side,
meaning those sites that are HTTPS have seen
no improvement at all, even though it’s a
minor signal, they’ve seen actually nothing at all in
terms of ranking improvements when switching to HTTPS. Do you have any
thoughts about that or– JOHN MUELLER: Well,
I don’t really know how they
checked all of this, but we do see this is
as a lightweight signal. It does affect– I think we
said 1% of the search results. So it’s not something that
is visible everywhere. So from that point
of view, it might be that they’re looking
at it slightly differently or slightly different. But it is something
that’s live at the moment. It does affect the
search results. AUDIENCE: Okay, thank you. AUDIENCE: That’s not
completely accurate, Barry. It only showed a not
scientifically significant increase when they
removed the outliers. For the outlier sites, there
was a major increase for HTTPS. But the outliers
where big sites like Walmart and stuff like that. AUDIENCE: No, the outliers
are big sites like Google. That’s a big outlier. JOHN MUELLER: It’s
really difficult to reverse engineer
this, considering that we do so many changes in
search all the time anyway. And trying to reverse
engineer one specific change and how it might be
affecting the search results is really hard to do. I think it’s interesting to
see these kind of reports, because it also something
that shows how webmasters or how these different data
analyst sites look at this data as well. So I think it’s interesting. I think it’s great that
people try to analyze this. But I don’t think
that you’d always see a one to one
matching with everything that we’d see on our side. But it’s definitely something
we do take into account. I think depending on what
you’re doing specifically on your website, what
is still left to do, you have to prioritize things. I wouldn’t necessarily
put HTTPS on the top of my list for all websites,
but for some websites it might make sense to put it
just a little bit higher up. For other websites,
it might make sense to put mobile a
little bit higher up. I think overall if
you’re a small business and you have a normal
desktop website and you can choose
between making a mobile friendly version
or getting everything moved to HTTPS, I definitely
recommend going with a mobile version
at the moment. There are a lot of wins with
having a mobile version. It’s something that
all users essentially see when they come with a
smartphone to your website. So there’s a lot of extra
possibilities there. But if you’re looking at a
complete redesign at some point and you’re saying,
well, I can put this in with my 50 other things that
I’m redesigning at the moment, then I’d definitely put
that in at some point. AUDIENCE: John, can I ask
you an important question regarding hreflang, please? JOHN MUELLER: Sure. AUDIENCE: One of the things,
obviously, you– [INAUDIBLE] and all these situations that
we’ve already been through. The thing that
becomes very apparent is that the site that
we have, a dot-com that has a Penguin penalty
is the X-Default. Now, I’m wondering now–
obviously everybody’s waiting for the update. Nobody sees it
coming anytime soon. We’ve got 130 offices worldwide. They’re affected by these
changes for years now. What would happen if
I created another site to be my X-Default? And kind of
relegated the dot-com being some obscure version
of whatever we can think is most convenient? JOHN MUELLER: It’s hard to say. So we’d probably respect
the hreflang markup in a case like that and try to
combine that with other signals that we have and use that
as the default page there. But it’s a complicated situation
in your case, because you have, I think, the UK version
and the dot-com version? AUDIENCE: Yes, that’s correct. UK has the hreflang, and the
dot-com is the X-Default. JOHN MUELLER:
Yeah, I don’t know. AUDIENCE: I don’t want
to have to do this, so– JOHN MUELLER: I
don’t know if there’s any– at least there’s no
clear, technical advice that I’d say go for either way. Because it’s possible that
we’ll just follow the X-Default. It’s possible that we just
have so many signals pointing at the dot-com version
that we’d say, well, this is more likely to
be the default version. It’s really harder
to say in general. Sorry. AUDIENCE: Yeah. I mean, I don’t want to go
down this route, obviously, because I’d much
more prefer that we know our site’s in
a good situation. Otherwise you wouldn’t
like our [INAUDIBLE] UK, which is ranking well. So I’ve kind of been backed
into a corner a little bit by the offices in different
countries who are struggling. Dot-com, we’re
still in 300th place for major keyturn
virtual office, but we’re in the
top five in the UK. So that’s such a disparity,
but the sites are identical. JOHN MUELLER: So I guess
one possibility might be to just do this for
individual countries or try it out with
individual pages. But I don’t think I’d have
any general advice and say, this is a really quick
way to fix this problem. But if you want to try it
out for individual countries, I could see that making sense. If you want to trust
the individual pages, that shouldn’t have an effect
on the rest of your site. AUDIENCE: And so from
an SEO point of view, the thing that obviously
I really don’t understand is that the [INAUDIBLE]
has no back links, but the dot-com
gets switched out. So I’m wondering if the
X-Default gets changed. Does the power to
the [INAUDIBLE] UK get affected by that? It’s a very difficult question
for me to even structure, but I hope you understand
what I’m talking about. JOHN MUELLER: I
don’t know for sure. I don’t know how
that would work out. It’s tricky because I
imagine most of the signals are pointing at your
main site, and it would be kind of
fighting with the signal that you give us and say,
well, actually my main site is this one. But I could totally
see this as something you might want to try
out on individual pages just to see how it works out. And I’d love to hear your
feedback on what you find out. AUDIENCE: Yeah, I’ll either
be applying for a new job or I hope it will go well. But yeah, thanks, John. JOHN MUELLER: All right. Let’s go through
these questions, because there are
lots of them there, and people submitted them. So, “my very non-spammy
design news website seems to have been
negatively impacted by the Google
algorithm Payday Loans. I’m baffled as to why
and what we can do. Any ideas?” If it’s not a
Payday Loan website, then it probably wasn’t affected
by the Payday Loans update, because that would be
something completely different. I don’t really see
anything specific with your site just
[INAUDIBLE] now. I think, to a large
part, it’s probably just ranking as
it normally would. But I can double check
after the Hangout and see if there’s something
specific I can add. “In our business, we rent
rooms in the same apartment. Each room has a separate page. Also, they have different
photos per page. Except the same
title and reference, most copy of the
original is the same. Do I need to set one of
the rooms as canonical?” You don’t need to set
them as canonical. It probably wouldn’t even make
sense to set them as canonical. It would be essentially
like having an image gallery and picking the first
image as the canonical. You’d probably want to have all
these pages indexed separately. So from that point
of view, I’d probably aim to have them
indexed separately. Maybe you can embed the
images all on one page, if that makes sense. But essentially, you
don’t need to set one page as a
canonical for something like an image gallery. “You stated last year
that Penguin [INAUDIBLE] should work on
fixing link profiles. Wait for refresh. It’s highly possible
no refresh is coming. Do you still stand by fixing
as the best way to resolve?” We are working on
a Penguin update, so I think that saying
there’s no refresh coming would be false. I don’t have any
specific timeline as to when this happened. Barry asked me jokingly
if it was happening today. And no, it’s not
happening today. But I know the team
is working on this and trying to find a solution
that generally refreshes a little bit faster
over the long run. It’s not happening
today, and we generally try not to give out too much
of a timeline ahead of time, because sometimes
things can still change. AUDIENCE: So faster
meaning that it’s going to be on a regular
basis, like Panda? JOHN MUELLER: We’ll see
what we can do there. So that’s something where we’re
trying to speed things up, because we see that this
is a bit of a problem. When webmasters want
to fix their problems, they actually go and
fix these issues. But our algorithms don’t reflect
that in a reasonable time, so that’s something
where it makes sense to try to improve the speed
of our algorithms overall. And some of you have
seen this firsthand. Others have worked
with, probably, other webmasters that
have this problem. And I think this is something
good to be working on there. AUDIENCE: But the impact
won’t be as big, right? Like before? JOHN MUELLER: That’s
always hard to say. And I imagine the impact
also depends on your website and whether or not
it’s affected or not. So if it’s your website, the
impact is always big, right? But it’s something–
we’re trying to find the right balance
there to make sure that we’re doing
the right things. But sometimes it doesn’t go
as quickly as we’d all like. AUDIENCE: Thank you. JOHN MUELLER: “Is it bad
to have links pointing to your site that are from
websites hit by Panda, Penguin, or a site with a
manual penalty?” No, not necessarily. So it’s not that
we’d look at the site and say, oh, this is a bad site. Any link from it is also
going to bring some badness to the site that’s
being linked to. But rather, if this is an
actual link, then that’s fine. If it’s not a natural link,
that’s not so helpful. There are some
situations where we see sites that are linking
out to all kinds of sites, and we might say, well, we can’t
really trust any of the links on the site, so we’re just
going to ignore the links out from this website. But that doesn’t have a
negative effect on your website. So if you’re being
linked to by a site that you suspect
Google is ignoring, then we’re just not passing any
page rank through that link, and it’s not negatively
affecting your site. It’s not having a
positive effect, either. AUDIENCE: John? JOHN MUELLER: I can’t
hear you, Joshua. [INAUDIBLE] AUDIENCE: All right. There, how’s that? JOHN MUELLER: Oh, it’s perfect. Great. AUDIENCE: These sites that
aren’t passing authority with page rank–
can we have any idea about– is that a small
percentage or a really big scale? I mean, in 20%, 30% or
more, or are we just talking about a relatively
small number of sites that don’t pass
on the authority? Because that seems to be
like a long term thing. AUDIENCE: It’s
something really, I’d say, fairly rare in
the sense that if we see that your site has
unnatural links from the site, we’ll also bring that
up in Webmaster Tools and use that as a manual
action in Webmaster Tools so that you’re aware
of the situation. But it’s not something
where I’d say a large part of the internet
is affected and essentially held back. It’s really only
for rare situations where we see that a large part
of the links from the site are actually unnatural links. Therefore we want to be
careful with any links that are coming out of this
website, just to make sure that we’re on the safe side. They’re not causing any problems
with the search results. AUDIENCE: All right, thanks. AUDIENCE: Hey John,
will you always give the manual action
for these kinds of sites? JOHN MUELLER: I think so, yes. So this is something
we do manually. It’s not something where
algorithmically the algorithm would just silently
go in and say, we’re ignoring everything
from this website. So since it’s a manual
action, we bring that up in Webmaster Tools
for this website and we’d let the webmaster
know about that as well. There might be some
extremely rare situations where something similar
happens on our logarithmic side or where maybe there’s some
manual action that we need to take that we can’t directly
show through the webmaster, but for the largest part, I’m
not aware of any exceptions at the moment. So this is something
that we would try to always show
to the webmaster. “I’m running a price
comparisons site. In July, we got a ranking drop
to the category and product pages, but the blog section
is still ranking well. So my question is, why is
the blog section ranking so well and the
other sections not?” I guess it depends a bit
on the type of content that we find there and
how we can differentiate between the content. And if you’re running a
price comparison site, and you’re just taking feeds
from various distributors or sites and essentially
just comparing them, then that’s not
that much value add, and that might be something
our algorithms look down upon. So if you’re not really
providing something significant, unique, and
compelling of your own, then it’s possible that
we won’t rank your site as high as it used
to in the past. And in the blog, maybe you’re
writing better articles there, and maybe that
looks like something that we prefer to show
in the search results. So maybe the ranking
will be just the same. But overall, this is
something– especially what price comparison sites
or generally affiliate based sites that you’d want
to watch out for and make sure that you’re really not
just combining these feeds and showing that one
to one on your site. You’re actually doing
something unique and compelling of your own on these sites. What makes sense for us to
send visitors to your site, because you have
more information about these individual
products, for example, then maybe even the
main website itself. So that’s something
where you really need to make sure that you’re
doing a little bit more than just combining feeds and
pushing them out onto the web. “If I change my site to
HTTPS and use a 301 redirect, do I also need to upload the
disavow file from the HTTP site?” I’d definitely also
upload the disavow file in cases like that. If you’re moving
from HTTP to HTTPS, or if you’re moving
into a different domain, all of these situations
I’d definitely make sure you’re
uploading the same disavow file so that we can take that
into account just the same. AUDIENCE: John, can I
ask a question, please? JOHN MUELLER: Sure. AUDIENCE: I have a big
doubt about some domains I want to buy, so I want to
ask you some technical things. Supposing I’m buying
some premium domain names from a company who
has something like 500, 600 premium domain names listed
online– also in the same web directory, same content, same
design, same outgoing links– basically, is the
same web directory published in all of these
premium domain names to keep them online
and live, OK? So my big problem is,
is possible, or better, what is the probability that
all these sites connected between them with links
to be penalized already by Google for duplicate
content or maybe by Penguin? Because they are all
connected between them. JOHN MUELLER: So we wouldn’t
do [INAUDIBLE] penalize it for duplicate content. We try to treat that
as a technical issue, and we might show only one
of these pages in search. But we wouldn’t penalize the
site for duplicate content. In general, if there’s a
manual action in place, you can fix that by having a
good website on that domain and doing a
reconsideration request. It’s important that there’s
something on that website, not just like an
empty domain name and you say, I want to
reconsider this website and I’ll put something
new up as soon as I have an OK from
the manual actions team. We really want to see
that there’s actually content there first. Otherwise it might
look like you’re just jumping back and forth
with the same content. So we really need to see
some legitimate content if there’s a manual action. Algorithmically, it’s a little
bit trickier in the sense that sometimes these domains
collect a lot of shady history. And that can take a little bit
of time to shake off again. So if you switch to something
that’s really legitimate, that’s a great
website, it can still take while for
algorithms to get rid of all of this bad, old
history and understand your new site the way it is now. So that’s something to keep
in mind where I definitely look at things like
to see what kind of content was on there before. If it’s a really blackout site
that was on there before, you see all these blackout links
pointing to the domain name, then chances are
those algorithms might still be affected
a little bit for a while. So if you need that domain
name as quickly as possible, you need to have something
visible in search. Maybe choose something
that has a little bit of [INAUDIBLE] in our history. If these were just directories
and they have no other history past that– maybe
the just have links between each other and no
spammy backlinks– then that’s something that
you can probably just put your new website
on and work from that. AUDIENCE: OK, so even if
they link between them, it should not the problem if
they are new site, new domain names? JOHN MUELLER: I can’t
say that for sure, but in general, if they don’t
have a really spammy history with spammy backlinks,
then I wouldn’t be that worried about it. We can’t give you this
information ahead of time, so it’s something– you have to
look into the history yourself and use your own
judgment into saying, it was just a part site. There was nothing
special about it. Or it was actually this
crazy blackout site that I don’t want
to be involved in. I don’t want to have to worry
about the history about. AUDIENCE: OK. Thank you very much. JOHN MUELLER: “Can I
get a manual action for great content, hyperlinks?” Wait. “Can I get a manual action
for great content hyperlinks pointing back at us that
are keyword anchorage?” So if someone else is
scraping your content and linking to your
website, in general that wouldn’t be a problem. That wouldn’t be
something I’d worry about. You wouldn’t get
manual action for that. That’s almost for sure, because
the website team manually has to look at these sites. And if this other site is
just scraping your content, then that’s not something
that we would say is going to be your problem. That’s more something where
we’d say, technically, we try to filter that out. If you want to take
action on your side, you could look at things like
a DMCA, which might apply, which might help you
to clean that up. But that isn’t something
where the manual web server team would take action
on your site for someone scraping your site. “We have indicated we prefer
the dot-dot-dot version of our site in Webmaster Tools. Should we upload the
disavow to both versions?” In general, we apply that to
the current canonical version. So if your website has
always been on dot-dot-dot, that’s the one I’d definitely
keep the disavow file on. If you’re switching
between these versions, I’d definitely upload
it to both versions just to make sure you’re
covering everything. “This week I saw an updated
site link search box. I don’t understand
the mechanism. Can you explain it?” So this is a specific markup
you could put on your pages where if we show a search
box in the site links section of the search
results– which we don’t always do for all sites–
but if we show that for your site for
specific queries, any searches that user is doing
there will be sent to your site to your site’s
internal search page. So for example, if you
have a WordPress blog, and you have search set
up, and you have the search box on your website, you add
the markup to your website. If a user in the search results
searches for something specific within your website, we
will redirect them directly to your website to
the appropriate search page within your website. So that makes it a
little bit easier for you to try to fine tune
your internal search results that you would show
the users and also to track how users
are clicking through into these search results pages. And this is a
fairly simple markup you can add to your pages. If you don’t have side
links shown for your pages, maybe it’s not worth the
effort at the moment. But especially if people are
seeing side links to your site, that might be
something that provides a little bit of extra value. “How is waiting determined
in the priority–” Oh, oops. It disappeared. I think the question
was about the ranking in the priority from
the crawl errors. We use, I think, four
or five factors there for the crawl errors priority. We have them in a blog
post where we initially announced this feature, which
was probably four or five years back now in the meantime. But it includes
things like, is this a page that’s linked internally? Is it included in
the site map file? Those kind of things. And if we can tell that this
is a more important page, we’ll show it a little bit
higher in the crawl errors section so that you see
that at first glance, and you can take action on that. If all the problem here as you
see in the crawl errors section are completely
random URLs that you think are totally irrelevant
to your website, then that means we haven’t
found anything that’s really useful on your website
that’s currently returning a crawl error, which
is usually a good sign. “Is there a negative
SEO effect when using a website builder
like Squarespace? I’m located in the Netherlands
and target Dutch customers. Squarespace web servers
are located in the VS.” I don’t know which
country that is. But, “What’s the
best choice SEO-wise when using Squarespace?” Oh, probably United States. So in general, we do
take the server location into account but only as a
very, very lower priority factor for geo targeting. So if you have a country code
top level domain like a .nl, .de, or you set the
location in Webmaster Tools, and we’ll definitely use that. And if we don’t have any
information at all about geo targeting for your
website, we might take the server’s
location into account. So if this is a good place
for you to host your website, go for that. That’s not something
I’d hold you back. That’s not something
that would have any kind of negative
effect with geo targeting. AUDIENCE: John, when
do domains– like I got an invite to get the
domains from you guys, and so, when is it going
to be available in Canada? JOHN MUELLER: I don’t know. You’d have to ask
the domains folks, but I imagine their answer’s
also– it kind of depends on– I don’t know what factors
they’re waiting for. But I know they’d love
to expand on that, but they need to make sure
that the base product is working really well first and
taking things step by step. AUDIENCE: Sounds good. AUDIENCE: Yeah, John, what is
the deal with all this stuff coming out in the US
and not for Canada? I find that unethical. JOHN MUELLER: Unethical. [INTERPOSING VOICES]
I don’t know, you guys set up those borders. I don’t know why
you have two states. It’s one continent, right? AUDIENCE: Justin Bieber– JOHN MUELLER: Seen you guys
just get along, you know? AUDIENCE: The US sneezes,
we get a cold, you know? JOHN MUELLER: I don’t know. I mean, in Europe, it’s
generally even later than Canada. So I’m sympathetic,
but I can’t really solve that problem
for you guys, sorry. AUDIENCE: It’s all right. Thanks. JOHN MUELLER: “Does
Google use the size of watermarks in image ranking? Do images with smaller
watermarks rank in image search better?” We do try to take the
quality of the image into account for
image search rankings. But usually if there’s a
reasonable watermark on it, that’s not going
to be a problem. If the watermark covers the
whole image, then of course that’s something that
changes the image from being a picture
about something to just being the watermark. But if you have a reasonable
watermark on there, that’s not going
to be a problem. So you don’t need
make it tiny, you don’t need to hide
it completely, you don’t need to remove it. Just make sure that the image
is still kind of the image that it’s supposed to be. All right, we have a couple
minutes left, so let me just open it up for you guys. What’s on your mind? I see the chat is
clicking away like crazy. What can I help you guys with? AUDIENCE: Can I– AUDIENCE: Oh, sorry, go ahead. AUDIENCE: Excuse me. Do you hear me, John? JOHN MUELLER: Yes. AUDIENCE: So returning
to my problem I asked before, if I
buy a domain name who has 500, 600 low quality
backlinks already on it, is that a problem? JOHN MUELLER: It
might be that these are links that are picked
up by the Penguin algorithm and where you’d need
to wait for a refresh. If you’re aware of
those spammy backlinks, I’d definitely at least
do a disavow for them what you get that site. But it’s not
something where I can guarantee that
it’ll go either way. If you know that it has this
kind of a spammy history, you have to work with
that and, I guess, take your chances if
you really, really want is that domain name. Or if you want to really
play it safe, you can say, I don’t have time to mess
with any bad history. Maybe choose a
different domain name. It depends a little bit
on how far you want to go, how much information
and knowledge you have about
handling these issues, how far you want to
fix this on your own, or how far you just want
to have something that’s as easy as possible and
is just like a new website from the beginning. AUDIENCE: You’re saying it’s
not the best choice I can have? JOHN MUELLER: Well, I think if
you’re very experienced, if you know about these
issues, if you can find all of these problematic
backlinks, and if you say, I don’t need this website
to rank from the first day, then maybe it’s worthwhile. Maybe it’s a really,
really good domain name that you might miss
out on otherwise. So that’s the two
sides of the equation that you have to think about,
whereas if this is just some random domain name that
you don’t really care about and you know it
has a bad history, then it’s probably not worth the
time and effort to fix that up. AUDIENCE: John, can I
ask you a recommendation? Would you recommend a
dedicated server as opposed to a shared server? I know it’s not
like [INAUDIBLE]. JOHN MUELLER: It’s whatever
works best for you. I guess it depends on the
website, how much resources you need, but you don’t need
to have a dedicated server. You don’t have to worry
about things like on a shared server in general, so
that’s not something where I’d say you
should go either way. It depends on your
website and your needs. AUDIENCE: Right, right. Because I heard that there’s
a– if your neighbor on a shared server does something
crazy– because I talked to this famous
hosting company there, they were saying
that hey, you can get penalized if a neighbor
does something crazy and you were right next door. JOHN MUELLER: It’s more
of a problem for us if the whole server
is spammy and you’re the only clean site on there. Then from a manual web
spam point of view, it might happen that
the website team says, oh, well, I found 50,000
domain names on the server. They’re all completely spammy. We’re going to take action
on this whole server. But if you have a normal
shared hosting environment, the host is going to take
care of those sites well. And they’re going to have
some kind of terms of service where they say, oh, well, if
you’re just spamming people, then we’re going to throw
you out of our system. And they kind of police
this a little bit. Then there are going to be
some spammers that get through, and a lot of normal sites
that are on the same server. So in those kind of situations,
that’s not something I’d worry about. It’s more, I guess, if you’re
buying your shared hosting from a known blackout that
is hosting all their blackout domains on the same server. It just wants to add some
variety into the mix. Then that’s probably not going
to be a good use of your time, of your money. But if you’re posting with a
normal provider, like, I don’t know, Go Daddy or any
of those other ones, then those shared servers
are going to work just as well as other ones. AUDIENCE: Thank you. AUDIENCE: John,
just a quick one. For probably the last
three or four months, I brought up a
particular website that is probably the
only spam result left in the results for my industry. And nothing’s been done
about it by anybody. And I presume by this point
that either Google or you agree with them, or we’ve
got the usual delay that we’ve seen for many,
many years of nothing being sorted out for it. And now they’re
ranking number one for almost everything
instead of employing the same link
building techniques. Should I open up a
webmaster thread about it? What should be done about it? But it’s bad for customers,
and it looks and reflect badly on the Google results. And the fact that I’m
bringing it directly to you multiple times, and you
either don’t agree with it all passed it on to
somebody else, also just goes to show that
we’re still not getting around the issue
of people like me helping to solve spam issues. JOHN MUELLER: Yeah, I’ll
double check with the analysts there to see what
they came up with. I don’t really know, exactly,
what the final situation was there, but I can
double check there to see if there’s something
I can bring back to you. AUDIENCE: I posted the URL in
the chat there, just in case you needed a reminder. JOHN MUELLER: Okay, great. AUDIENCE: Thanks, John. JOHN MUELLER: OK, and with
that, we’re out of time. So thank you all for all of your
time for the good questions, for the good feedback, and I
wish you guys a great week. AUDIENCE: Thanks, John. Have a great week. JOHN MUELLER: Bye, everyone. AUDIENCE: Thanks, John. Bye. Thanks.

Leave a Reply

Your email address will not be published. Required fields are marked *