Reconsideration Requests
Show Video

Google+ Hangouts - Office Hours - 09 May 2014

Direct link to this YouTube Video »

Key Questions Below

All questions have Show Video links that will fast forward to the appropriate place in the video.

Q1: If the cache date for a page that I have an external bad link on is more recent than my disavow file, Can I be certain it has now been disavowed? | Show Video

A: Yes, but it if the cache date is not newer this does not mean that it has not been crawled. It is also possible that it has been crawled but not indexed.

My Thoughts: What John is saying is that if you use the domain: directive then every time a page is crawled on that domain it will apply the disavow to that page. A page must be crawled for its status to be changed for your url. It is not applied to the whole website at once, instead on a page by page basis when it is crawled.

Q2: Does Google use the content from your smartphone version of your site or the desktop version to calculate ranking positions for smartphone queries? | Show Video

A: It uses the desktop version as that is the one that has the most content and then swaps out the URL for the smartphone version.

My Thoughts: This works in a similar way for reactive designs, however there is no need for Google to swap out the URL as it would be the same. However you if you have a smartphone design Google may favor you in SERPS over others that do not.

Q3: Why do my rich snippets not show for my website in Google? | Show Video


My Thoughts: John provides a great in depth guide at this part of the video

Q4: How do you stop bad links from press releases that you have no control over? | Show Video

A: If you do not do this on a regular bases then do not worry about it. But if you are concerned then use the disavow file.

My Thoughts: This conversation was very interesting as I ask two follow questions about naked links that low quality sites turn into dofollow links when they copy the content and how that affects the Penguin algorithm. John says the algorithm takes into account the sites backlink profile as a whole and at that point it negatively suppresses a site if this is a tactic that is used often. So if you do press releases infrequently you are unlikely to have a bad backlink profile thus no suppression is applied.

Q5: Once a manual penalty is lifted, do you rank naturally? | Show Video

A: Yes & No. You must wait for a refresh to see your site return to where it would normally rank if all things were back on a level playing field.

My Thoughts: A Must Watch! This is a perfect example of how the correct wording being used is critical in getting the right answer so that you understand what is going on. Essentially John says that they call natural rankings anything that we see in SERPS right now! However a website owner would not see it that way. Neither would anyone else looking at this from an outside perspective. Every algorithm update only shows natural listings. if you have a manual penalty then you also in some way would be naturally ranked according to your penalty. However it would be impossible to have your pages rank where they would if you had no previous manual penalty until the Penguin algorithm refreshes and puts you back on a level playing field. This is probably one of the most insightful question/answer we have seen related to Penguin to date.

Q6: Can you have algorithmic issues with links that are not in your Webmaster Tools links list? | Show Video

A: Theoretically it is possible, but mostly the links not shown are just more from the same domains that are listed. You could have a ton of small sights that all only have a link or two and then it would be difficult to show them all but it is extremely unlikely. The links list will show the most problematic links always and those links are primarily what Google uses for the algorithm and manual actions.

My Thoughts: So cleaning up links from domains that are not listed in WMT links list are not going to have much or any affect basically.

Q7: Do keywords in the url help with rankings? | Show Video

A: Yes, it is taken into account slightly.

My Thoughts: With all sites being equal it may be that one thing that puts you above the competition but it is a very small factor.

Transcript Of The Office Hours Hangout
Click on any line of text to go to that point in the video

JOHN MUELLER: OK. Welcome, everyone, to today's Google Webmaster Central Office Hours Hangout. My name is John Mueller. I am a webmaster trends analyst at Google in Switzerland. So I worked together with engineers, and together with webmasters, like you guys, to help make sure that communications flows in both ways. Looks like we have a bunch of people here already. We have a lot of questions that were submitted already here, as well. You're welcome to add more questions if you want. If you're in here live, feel free to ask away during the answers, or the questions if you have more comments or ideas. And as always, I think there are a few more events lined up where you can also add your questions. So if I don't get your question here, feel free to add it to one of the future ones. All right, before we get started on the questions, do any of you want to ask the first question?

MALE SPEAKER 1: Hey John. I'd like to come in with a question. Thank you. I've noticed that Google is trying to access some 404 resources for up to three months plus. I mean, it's like indefinite. It just keeps coming over and over. And it still gets 404 as those resources, whenever I return to the website [INAUDIBLE]. And they are quite a lot. Based on the last of the videos posted by Matt Cutts, in which he said, we can use 410 as an alternative. What happens if I replace a 404 with 410 [INAUDIBLE] [INAUDIBLE] to try to save some server resources?

JOHN MUELLER: I think you might see a small change in how we crawl them, but probably not something big. So if we're crawling, usually if we recognize that this is a 404, if it's been a 404 for a while, we won't crawl it as frequently as we would a normal page. So that might be instead of every week, maybe we'll crawl it every month. But of course, if you have a lot of these URLs, if you have things like session IDs, or something like that that always return 404 now. Even if we crawled the individual URLs once a month, that could still be a lot of requests. So I think what would happen with it 410 there is maybe we'll crawl a little bit less frequently than a month. But even if that's every one half months, that's still something that could add up if you have a lot, like a large number of these URLs. If the crawling is effectively a problem there, you could use a robust text directed to block crawling of those URLs. But past that, there's no way to really prevent us from trying those URLs again. Because we think, maybe, OK, fine. There will be a link to those pages. Maybe if we know about these pages and think, oh, maybe there's good content here now. We'll just want to retry them to double check that we're not missing anything important. Yeah. Go ahead.

MALE SPEAKER 1: I wanted to tell you about, we encounter this kind of coming back to the returns. For real estate properties. A real estate property, when it's sold, I mean, it won't be placed to resell again any time soon, you know. And there are a lot. There are tens of thousands of real estate properties which gets into 404. I mean on a daily basis. So we thought that if we could change 404 to 410, maybe Google will definitely understand that these properties are no longer available, as they were sold.

JOHN MUELLER: We'll still want to double check them. Especially if we find a new link to those pages. We'll want to double check them from time to time. What really the main difference between a 404 and a 410 is kind of the time it takes from URL that is indexed currently, until it's dropped out of the index. So a 404 is something we'd see as being potentially temporary, in that, if we know of a page, and it turns into a 404, we'll retry it a few times before we remove it completely. Whereas with a 410, if we see a 410, for a page like that, we'll drop it a lot faster. So the change will be faster. But afterwards, we probably won't necessarily be less frequent.

MALE SPEAKER 1: I understand. OK. Thank you very much. Then perhaps we can try with one website to see how it goes. And then we will decide. The only problem in my head, it was not to go on the other side to get something wrong by testing it. Can I step in with another short question?


MALE SPEAKER 1: I've seen related also with often crawling that the old directive crawl delay, from the robots.txt got banned by Google. I mean, it gets ignored. And in Webmaster Tools we even get a message saying that Google doesn't support any more of this directive. How can we delay the crawl? I mean, Google comes in crawl in a very big amount, at certain points. And if we can make this delay, which we used to use as crawl delay parameter, it will be a lot easier for a lot of websites.

JOHN MUELLER: We actually never supported that directive. So I think that's something from Yahoo, specifically. But we never supported that. So the biggest problem we have with that directive is that the values specified there were completely unusable. And that people would say, you can crawl five times a day. And for us, crawling five URLs a day for a website is not going to get us anywhere. So that's essentially why we don't use that. What you can do in Webmaster Tools, you can set the crawl rate for almost all websites. A maximum crawl rate. For websites that don't let you set a crawl rate, you can use the Report a Problem with Google Platform, which is there as well. Which lets you send a report to the team working on the Googlebot to let you give them feedback on the crawling, where you could say, oh, you're crawling much too much. Or maybe, if you could crawl more. If you crawl during my evening times, when my users aren't on the website. So that's the kind of feed back you could give the Googlebot team there.

MALE SPEAKER 1: Oh, I see. Maybe, as a suggestion, maybe you guys can implement something like a crawl rate. I mean, if it will be in seconds and if we can set up a delay. It will be much easier for larger database of websites. I mean, if we manage a lot of websites.

JOHN MUELLER: Yeah. I mean, for most of them, you can set that directly in Webmaster Tools. What might make sense, at some point, is putting that into the API. And I know it was in the old API. So that you can programmatically set that if you wanted to. Maybe that would work in your case?


JOHN MUELLER: All right. OK. Let's run through the questions that were submitted and see how far we can get there. Does Googlebot or Google Letters send incorrectly formatted no file requests? For example, if someone writes realm of follows, row equals no space follow, with a space or with a dash, will Google understand these? No. We don't actually understand these. So one tricky part there with these algorithms is if you have spaces in them, they're actually different directives. So they're separate directives. And with a dash, that's something we don't support as well. So if you really want to no follow a link you really need to do it technically correct. And you don't need to use crawl. So you could just use rel equals no follow. Or use quotes if you want to. But essentially, the no follow has to be written there correctly, and has to be clearly a part of the link. Recently, number of duplicate title and description tags have been increased by large numbers. And webmaster reports the reason for saying is that Google is indexing URLs with position IDs, query parameters, even though the canonical has been set up. Why is this happening? From a technical point of view, what happens when we see a URL with a rel canonical is first, we have to index that URL, so that we can actually look at the content of that page. So if you have a URL that you don't want to have canonical, and we find it and we crawl it, we find the rel canonical, firstly it's a version that actually you supplied like that. And then when we go through the content of that page, we'll find the rel canoncial. Then we'll go off and crawl the other URL and see if they match. And then pull them together. But the first step is really to index the URL as you have it there. So my recommendation there would be if you consistently have this within your website, if you have a large number of URLs that are being crawled and indexed even though they're not canonical, I'd try to work on making sure that at least internally within your website, you're linking consistently directly to your canonicals. So as much as possible, instead of providing add URLs and then using the canonical to clean that up, make sure you're providing the right URLs from the start, so that we don't have to go through this canonical change. That said, this is the information, just for your information, in Webmaster Tools. It doesn't mean that this is going to be a problem. It doesn't mean that this is going to cause any issues with your crawling, indexing, or ranking of those URLs. This is essentially just providing you with this information so that if you weren't aware of that, now you have that information. So it's definitely not a critical thing. But if you want to clean it up, cleaning it up on your website directly without relying on the canonical to clean that up for you probably makes sense. What do you suggest for e-commerce content strategy? Wow. That's a big question. I probably need to look into that a little bit more what specifically you're trying to find out there. And maybe one thing you can do is start maybe a thread in one of the help forums or on Google+, and get advice from other webmasters. And kind of elaborate a little bit more of what you're looking at. Because making e-commerce signs is a really big topic and there are lots of possibilities. And somethings work well for some websites and other things work well for other websites. So there's no one size that fits all of these sites. If our design requires our page header to be an image, do we lose any SEO or AdWords quality score, variables,

benefits to implement an image with alt in our h:form tag instead of text? I don't know how AdWords handles these kinds of situation. I can't speak for them, so from that point of view, I can't really say. With regards to search engines, one of the things to keep in mind is that we tend not to look at the text that's within the image. So if this image, for example, is a heading that explains what your website is about, what your business is about, and this is text within that image, then that's not something that we're going to be easily able to take into account for your website there, for those pages. So if this is your heading, it's all in an image file there's, no alt text, maybe there is no other information about this on your page, then that's really going to be tricky for us to actually pick up. So from that point of view, having a clear heading on your page that's readable in text, as something that explains what your page is about, explains what your business is about, usually makes sense. If you're talking about a logo, for example, where you have your company name. But you also have your company name in the alt text. You also have your company name elsewhere on the page, then usually that's less of a problem. We have enough information about your pages, about your company, to understand that this is a part of your page there. Whereas, if this is really the main heading of a page, and it's all in an image, then that's really tricky for us to pick up. We give out EyeFrames to affiliates. But sometimes affiliates don't embed the EyeFrame properly. Let's take a quick look at what kind of EyeFrame this is. What we're looking at. Whoops. And I need to log in seperately for that. OK. So I can take a look at the details there. Sometimes affiliates don't embed the EyeFrame properly, but link to the code. The image on top is just seen as a link. Should we try to stop this? Let me see if I can get the URL anyway to see if we can find something there. Let me take a quick look. Oops. Too many flipping tabs. I shouldn't play around. It's so much fun in the Hangout. I'd have to take a look at the details there. That's really hard to say without looking into the details. There's too much. So what I'd recommend doing is maybe starting a thread in the help forum. Maybe starting something on Google+. And just adding me there. And I'm happy to take a quick look to see if there's something specific that might be problematic there. If I added a site to my disavow file on, say, March 1st, and I see the site has then cached on March 7th, then can I assume that my disavow for that domain has taken effect? If not, is there a way to know that a domain that I've disavowed has actually been disavowed? The disavow file, if you do a domain directive into a disavow file, essentially that means every time we crawl a page from that domain, we'll take that disavow file into account. So this isn't something that's based on the homepage of that domain because the homepage might change differently than actual pages on the site. So it would be something that's more based on the individual pages themselves and the links from there. So, in a case like this, if the link was from the homepage, and you check the cache page and you see that it's updated, then it definitely that's been taken into account. If this is on an individual page of that site, and you've seen that it's been updated in the cache, then that's also definitely been taken into account. In general, I wouldn't use the cache page as the main method for recognizing when a page has been update in our index, because a lot of times we'll update our index without updating the cache page. But if you do see an update in the cache page, then we've definitely updated our index in the mean time. So the cache page is kind of like a way of double checking that, if you wanted to check that individually. But it's not a guarantee that if it doesn't update, that it hasn't been crawled and indexed again. And again, this is on a per page basis, so not on the domain. It's not the case that if we recache the homepage of the domain that we've recached everything from that domain.

GARY LEE: John, are there any other ways that we can check that? Or is that the only kind of sort of half baked method we can do that?

JOHN MUELLER: That we've crawled individual pages from site. I guess the cache page is a reasonable way of doing that. And again, it's possible that we've updated our index without updating the cache page. But if the cache page is updated, then we've definitely updated our index in the meantime. The tricky part, I guess there, is if you have a lot of pages on a website, then it's going to be hard to double check if all of these were actually updated. But the cache page is certainly a way of seeing if the index has been updated. It's not a guarantee, if it doesn't change, that it hasn't been updated, though.

GARY LEE: There's no other sort of techniques we could use to dig out information otherwise. That's pretty much the only avenue that we would have to go down.

JOHN MUELLER: Yeah. I mean, some pages have things like a date stamp on them. And what you can do sometimes is do something like a site query and search for a specific date to see if that's been picked up. That's something I like to use. It works really well, if these pages have a date stamp on them. I tend to use them, just to double check that we're crawling properly. And to check, for example, international time websites. But a lot of sites have a date stamp on them and you can find things like that to kind of double check that they were actually picked up.

GARY LEE: OK. Thanks, John.

JOHN MUELLER: For purposes of ranking on mobile search, does Google use content from desktop or mobile site, in the case of a mobile site that has less content than a desktop site? The site in question has responsive design on one URL in both desktop and mobile. For both desktop and mobile. So usually what we recommend for smartphone websites, I'm assuming this is for smartphone websites, is that you use a rel canonical to point to your desktop page. So in a case like that, we'll recognize the connection betweed the desktop version and the mobile version. but the version that we actually use for indexing is going to be the desktop version. Because, like you mentioned, usually that's the one that has the most content. That has the most internal links, as well. So we can crawl the rest of the site. So what would happen there is if you set it up like that, we'll crawl and index both of these versions. We'll focus on the desktop version to show in search. And we might swap out the URL for the mobile version for users of smartphones, but we'll actually use the content of the desktop version to actually do the ranking. This can, theoretically, get a bit complicated if your mobile version is actually the one that has more content than the desktop version. Which, theoretically, could happen, if you have something that's really focused on mobile users that doesn't make so much sense for desktop users. And, in a case like that, we'd recommend not using the rel canonical between these two versions because you essentially want your mobile version to be ranking independently of the desktop version in a case like that. But in most cases, where the mobile version has less content, where there's a clear connection between the desktop version and the mobile version, you'll want the desktop version to be the one used for crawling, indexing, and ranking. Mostly.

GARY LEE: So John, is it possible, then, that you might have your mobile site, your smartphone version, and somebody does a query search for explanation of how widget is used, but when you get to the mobile version, that kind of explanation is taken out because obviously, it's too big for the mobile version. Would you still deliver [INAUDIBLE] on their cell phone to that mobile version? Or would you then deliver them to the desktop version? Even though, I guess when the customer lands on the page, it's going to be responsive and turn into the mobile version, anyway.

JOHN MUELLER: Yeah. So from our point of view, it's important that the primary content is equivalent. And usually, that's not a problem because on the mobile version, you'll drop things like the sidebars. Simplify the menu. Simplify the footer. Maybe reduce the size of the images. But the primary content is usually the same. So that's essentially what we look for with that connection to the mobile version, to kind of confirm that we have it right. But if the mobile version really doesn't have the content of the desktop version, and we send them to the mobile version, or the site itself redirects the user to the mobile version, then that's kind of a tricky situation that we can't always catch. But we try to recognize those situations and make sure we're not associating with them appropriately.

GARY LEE: All right, thanks John.

JOHN MUELLER: OK. We've recently added some pagination to our site so that Google can crawl and pick up more pages. However, the index page count has hardly moved. Shouldn't I be worried? Generally, what happens with pagination, if you're talking about rel next, rel previous, those kind of things, then that's something that helps us to better understand the individual pages of the site. But we wouldn't necessarily crawl and index much more content there. Because you'd still kind of really need those normal HTML links between the individual pages anyway. So that we can crawl and index those individual pages. So, In a case like this, if you're adding pagination buy using rel next, rel previous, then that's something that you'd more likely see some of the results in search. So in your ranking for those pages. In which pages would show in the search results. Then you would for the number of indexed URLs. If the number of indexed URLs is a problem, then usually that's a sign that the links between these pages aren't really working out that well. Or, perhaps, that we really know if we can trust this website enough to actually spend so much more time picking up all of the content from those pages. So what I'd do there is primarily double check the technical situation to see that we're actually picking out all of the content, that we're able to find the links, and then, work on making sure that the quality of your content is really the highest it could be so that it's not just a matter of putting content out there that you want to have indexed, but actually that it's high-quality content, that it's something unique, something compelling to the user, something we should be showing in Search, something that's not just aggregated or copied, or rewritten from other websites. It seems like there's a flag penalty, which prevents rich snippets from appearing for a site. If it's true, what can I do about it? This site used to have stars, now, not anymore. There are few things that can influence whether or not we show rich snippets for a website. I think we have a pretty good Help Center article on that. The primary things that you'd want to watch out for are the technical issues, whether or not they're implemented correctly. Then, of course, their policy questions, whether or not they're implemented in a way that we want to show to users. For example, on homepages it almost never makes sense to show stars, review rich snippets, because your homepage isn't about one specific product. It's usually about your company, about something general. So that's not something where you'd want to show those rich snippets. Other kinds of policy issues, for example, is if those reviews aren't actually reviews by your users, but rather reviews that you're aggregating from elsewhere. So if you're taking a feed from a big e-commerce site. If you're taking a feed from some review website and just showing those on your pages, then that's not something we'd show in Search as a rich snippet. Similarly, if the reviews on your pages are kind of about your content and not about a product or a service that you're describing, then that's something we also wouldn't show in search results. So we've see this, for example, with plug-ins that are available for WordPress where you can just add the five star review thing to your blog, and your readers can rate your blog posts with those five stars. And your site kind of shows those as rich snippets to Google that that's something we wouldn't want to trust. Because essentially, it's not a review of a product, but rather a review of the text that you're providing there. That's not something we'd want to show. So those are the policy questions that are involved there. From a kind of higher level point of view, we also take a look at the quality of a website overall. So if we find that we can't completely trust this website with what it's providing with the quality of its content, then we're much less likely to actually show rich snippets in Search even if they're implemented correctly. Just because if we can't really trust this content, why would we want to show it even more visibly in the search results if we already know that maybe there's something problematic with this content. So that's the more subtle kind of issues that you might be looking at there. What I'd recommend doing is taking that and posting in one of the Webmaster forums and asking some of the more experienced users there for their opinion, on the one hand, about the quality of your website, on the other hand, about things like policy questions with regards to your use of the rich snippets just to make sure that you're covering all of the possible issues there. I'm assuming if your snippets used to show up, then technically they're implemented correctly, so that should be OK. You can double check that in our testing tool, though. So that's kind of the baseline that you would want to work on and then making sure you have the policy and the general quality issues covered as well. Now, what do we do if we submit our site news in a good authority news PR site, but it consists of [INAUDIBLE] poor authority news sites, which may be done automatically, and our site gets a poor authority backlink. Will Google consider it unnatural, and we need to disavow it? So generally speaking, you should've taken care if you're just submitting articles, for example, to article databases or PR websites if you're doing that primarily in the hope of getting a link back from those sites. Because that would be considered a natural backlink regardless of the quality of the site that's actually linking to you. So essentially, if you're providing this article for them, and in exchange for them being able to use your content they have to your site, then that's considered a natural backlink, and that doesn't really matter about the quality of the site itself, but essentially just the whole process there. On the other hand, if you're submitting a PR article in the sense that you're kind of explaining your new service, explaining your new company, explaining a new product that you have, and this is something that you're putting out on PR websites and you're not doing this to get a linkback. Maybe you just have your URL in the footer somewhere, it's not even linked, or it has a rel="nofollow" then generally, that's fine. And that's not something where I'd worry about lower quality websites scraping that, and reusing that, but essentially, instead of looking at the quality of those websites, take a step back and think about why you're doing this. Are you doing this to kind of get the word out about your product. In that case, I think a nofollow link on there, for example, would be absolutely fine. If you're getting the word out people can go to your website to check it out. On the other hand, if you're doing this primarily to get links, then regardless of the site that's publishing it, that's going to be an unnatural link and something you might want to take care of.

GARY LEE: John, if you're doing it for legitimate reasons and it's a naked link at the bottom of a PR site of some kind, and then a scraper decides to pick that up, and their design of that site is to take a naked link and make it a clickable link, which many of these really low-quality sites actually do, then you don't really have any control. You're not putting a link in the article in the first place. You're simply putting a naked URL as the reference. And then the other websites are responsible for picking up your content and doing things with it that you had never intended. And I think that's the question that's circulating around many forums and places at the moment. Is there an answer for that?

JOHN MUELLER: So I think there is a two-part answer for that. On the one hand, we try to recognize these as much as possible and just ignore them automatically. I think we're pretty good at that. And the other part is if you do recognize this yourself and if you're aware of the problematic behind it, you could put it in your disavow file and just forget about it. Just put the whole domain in there if you know this is a really low-quality PR site, and you never have to worry about that again. So you can take the problem in your own hands and take it out and make sure it doesn't cause any problems. Or if you don't know about this, and this is something that essentially just happens for one of these PR releases that you did, and your website isn't normally connected to the rest of the internet, and you're not out buying links anyway, then this is something you generally don't need to worry about anyway. So we do take a look at the bigger picture as well. We take a look at the bigger picture algorithmically as well, and if we see that you're not pushing out press releases every couple of days with links in there, then that's something that's not going to cause any problems.

GARY LEE: When you say that, do you mean that algorithmically or manually? Because obviously, the algorithm isn't controlled in that sense. The algorithm picks up what it picks up. And if it's nasty then you kind of have a suppression of some sorts.

JOHN MUELLER: We do take that into account both algorithmically and manually. So we try to look at the bigger picture and see how things are working for this website and if it looks like the website is primarily using these kind of methods for link spam, then the algorithms will pick up a lot stronger on that, whereas if they see this happening once or twice, or a couple times, and the rest of the website is doing normally, then that's something the algorithms will say, well, weird sites link in weird ways. The Webmaster can't control all of that. We shouldn't penalize this website for something that some random spammer is doing that doesn't even know what they're running on their servers.

GARY LEE: I think that's some great information there. I think that will give a lot of people a lot of confidence actually. And that's something I think we've been waiting to hear. So that really is a great bit of info from this Hangout from France.

JOHN MUELLER: Thanks. Alright. Here's one about coupon sites. I'm running a coupon site, which is providing discount coupons and deals from Indian shopping stores. My question is what should I do with my store page, which don't have any coupons for some time? I think usually when it comes to coupon sites, I think there are a few things that are always problematic or tricky to handle there. On the one hand, it's important that you really have unique and compelling content of your own on these sites so that when users go there they actually find something that they can use. So things like outdated content that they can't use anymore, content which isn't valid, the scraped content from other sites, descriptions, images, those kinds of things scraped from other sites. That's all content where when we recognize that this is happening on a website, our quality algorithms will go and say, I don't know if we can really trust this website that much. And they might take a look at the website overall and see there's a large part of the site that's kind of low quality, or that could be seen as low quality. And that's something that can affect the whole website. So that's, I think, the primary problem I see with regards to these coupon sites. Your question, I think, with regards to store pages that don't have any coupons for some time, I would tend to see that as an empty search results page. So someone is searching for a specific store and there's no content that you have available for that store, then essentially that's an empty search results page, which is a bad user experience. So what I would do there is make sure that those pages aren't actually indexed like that. So one thing you could do is, for example, put a noindex on those pages so that we can crawl those pages, we can see if there's content there. We can see the noindex and say, OK, well, the Webmaster doesn't want this index, so we won't index it. And if we don't index it and don't show it in Search, then from our point of view, that's fine. I mean, that's not something where you're taking low-quality content and presenting it to users in Search. The content they find in Search is useful for them. So if you know that you have these pages on your site that don't actually have any content on them, make sure that they're not indexable, so put a noindex on them. In those cases, people who are still using your website and searching within your website, they can still go there. They can see that this store is tracked by you website, for example, but that you don't have any content for it at the moment. So they can still interact with your website in that regard, but those empty pages don't actually show up in Search. And it's fine to have pages that fluctuate between being indexed and having a noindex. So if you have content for them now, and you don't have content for them next week, then put in a noindex next week, and maybe a week afterwards you do have more content, taking that noindex away is fine. And that's something you can probably even automate in a way that you don't really have to worry about it individually. The ALT text from an image on a website, will it appear as a text snippet in Google Search? Sometimes. So we do take the ALT text into account when we index pages for a web search. So that could potentially show up in the text snippet of a result as well. I have a website, I read the news, I write the posts on my blog, and some key words match, but I still give credits to the content, like, if it is about a Google Doodle, I'll give credits to the Google Doodle. Is it copyrighted stuff? So I think the general question here is you read about something in the news, you find out about something, maybe online, and you write a blog post about it essentially rewording what you found online. And from our point of view, that quickly gets into the low-quality content area in the sense that you're taking content that already exists online, you're changing it around a little bit, maybe rewriting it in your own words, and putting it back online. You're essentially not writing anything new while doing this. You're kind of just rehashing the stuff that's already found there. And providing a link back or even giving credit to the original source isn't really going to make your content higher quality. So what I would do in a case like this is maybe find a niche where you're really comfortable in. Where you know a little bit more than the average reporter out there. Where you can take a news report that's online somewhere and turn that into something that has a little bit more value. Or take, maybe, multiple news reports and compile the unique information from there into something that has a little bit more value on your side. So instead of just taking content from other sites, rehashing that, and putting that on your website, make sure you're providing something that makes sense to crawl and index of your own so that when Google comes along and sees, OK, I have this original source, and I have your rewritten copy here, our algorithms can quickly say, well, both of these are actually good pieces of information. We could point the user to both of them. Whereas, if our algorithms are looking at it and saying, oh, well, this has really good information written by a high-quality known author, and this one over here is kind of the same as this one here, so we don't really need to index it separately. And our algorithms, when they look at a site overall, they might say, oh, well, a large part of this site here is actually just rehashed content from this site here, or maybe from two or three other sources. So it doesn't really make sense for us to even put that much effort into crawling and indexing this site here, because it's nothing really unique, nothing interesting of its own. So from that point of view, I'd really make sure that what you're providing actually provides more value than just a copy of the content. And there are lots of ways that you could go out that, including things like having a lot of users who are active on your site who are discussing this topic as well. So user-generated content is definitely something you can build on as well. But just taking content, rewriting it, and putting it on your site, that's probably not a good strategy for the long run.

GARY LEE: John, can I ask you a question, please?


GARY LEE: So we implement the hreflang and [INAUDIBLE] day our website's up. And we appear to be ranking reasonably well considering we have no backlinks on the site and various other things. And ranking much better than our, which is still suffering. Which would suggest that our content is of a reasonable nature and that people like something that we're doing. So it would suggest that our is under some form of suppression, which we are aware of. Now, what else can I do now? Am I in a waiting game for this next Penguin refresh, or, I mean, I went in and I managed to disavow another 300 links that Google Webmaster Tools didn't find, but some other programs did. But I'm out of ideas now. Is there anything else that I can do?

JOHN MUELLER: I'd have to take a look at your site specifically there. But in general, if, for example, an algorithm like Penguin has picked up issues in the past, then that's something you kind of need to wait for that algorithm to update. So there's no work around around that algorithm to force it to update or to be whitelisted by that algorithm. That's something that really needs to happen naturally.

GARY LEE: But without a refresh, that wouldn't happen?

JOHN MUELLER: Yeah, exactly.

GARY LEE: Yeah. OK. Cause, I mean, there have been various discussions even on your Google + page. So there was one where we were discussing whether or not-- I think it was Allison from the [? pop-up ?] computer on Google Forum, she was saying that you rank naturally where you rank now. And very clearly that isn't the case. Especially seeing the [INAUDIBLE] UK site live that that isn't the case. There is algorithmic issues. And as you said, we do have to wait for a refresh.

JOHN MUELLER: Yeah. But I mean, it's hard to say whether or not-- I mean, it's almost, I guess, arguing semantics with regards to ranking naturally, because, at the moment, the site is ranking naturally if it doesn't have any manual actions on it that are manually pulling the site down. And if these are algorithms that just haven't been refreshed or had their data updated, then that's still, from our point of view, a part of the natural ranking. But I also see your point of view in the sense that if this algorithm was updated then it would show, perhaps, different data. But that would be the new natural ranking at that point.

GARY LEE: Yeah, I think anybody looking at it from my point of view would see that as a suppression. And it is semantics. It is a case of wording unfortunately. I think, yeah, it is a very difficult thing to look at on that front. And one thing that stemmed from that that we spoke about last week or the week before is I'm seeing a very large amount of traffic and URLs of people coming to our site, but titles that are incorrect even though we are right. Even though we've got them written correctly. So we're getting people, we're getting phone calls from people asking us for a service that we don't provide because titles have been rewritten.

JOHN MUELLER: Could you send me some examples of that?

GARY LEE: I will do.

JOHN MUELLER: So some search results where you're saying you searched for this keyword and Google is showing this title, which is clearly wrong. Because that's just the kind of thing we can take back to the titles team, and they're very receptive for this kind of feedback because we work hard on making these titles work. They have to fit the limited space that we have. The space is limited depending on the device as well. So that's something where if we can improve those algorithms then it's not just your website that sees an improvement, we will be able to improve that across the board.

GARY LEE: Yeah, I'll send you an email right after this and [INAUDIBLE].

JOHN MUELLER: OK. Thank you. All right. Google decided this was thin content and placed the manual action on a site. We immediately placed a meta noindex, follow on all geographical pages. Google dropped about 30,000 pages from its index pretty quickly. However, about 2,600 pages-- let's see, hard to tell which ones fit together here. Let's see, maybe this one-- about 2,600 pages which are being purged from Google's index very slowly. The issue has been going on for three months and we're doing everything we can to show Google we're good guys. We put in a reconsideration request almost 29 days ago, but heard nothing back-- as yet good or bad. And I put in another reconsideration request two days ago, so how can we have the manual action lifted, or if they split that, which is the best way to remove the remaining pages to satisfy Google of our intent? These are long combinations, hard to figure out which ones belong together. Leave the meta noindex in place, or place of what had gone on these pages, or something else. So I didn't quite see where this started, what essentially happened with your website. But essentially, if you know that these are thin content pages, then putting a noindex, follow on there is absolutely fine. The follow is the default behavior, so just putting a noindex on there would be just as good. Putting a 410 Gone on those pages would also work from our point of view. The difference between the 410 and the noindex is primarily on the user side, though. So from our point of view, both the noindex and the 410 would result in that page dropping out of our index. From a user side, a page with a noindex is still accessible, still normal to be used within a website, whereas a page with a 410 usually is something that would show an error to them. So if this is content that you think is thin but not necessarily thin in the sense that you don't want users to see it at all, then maybe noindex is better. Users can still use your website to find that content, they just won't find it in search. Whereas if it's a 410, then users wouldn't be able to go to that content at all within your website. So depending on what you're trying to do there on the user's side, you could choose between the 410 or the noindex. From our point of view, both of them are essentially equivalent. I'm not sure which kind of manual action your website initially had there, so it's really hard to say what you should be focusing on there. If this is something where you received a notice saying that the quality of your content is bad or it's thin content, those kind of things, then cleaning it up like this is pretty good on the one hand. It's a really strong sign that you're trying to do the right thing there. It really kind of depends on the rest of your content, though, with regards to how the web spam would look at that. For example, if you're aggregating news from a variety of sources and you're just noindexing the pages that don't have any content at all, then that's probably not going to be enough. They probably want to see that you actually have something unique and compelling of your own where we could send users without having to worry about your website. On the other hand, if it was only these pages that were actually the problem, then using the noindex there is probably OK, and you probably know yourself whether or not the rest of your website is something that is of high quality that provides a service that can't be found elsewhere. What you could also do if you'd like me to take a quick look at that is send me a short note on Google+ and I can take a quick look at that with the web spam team. I can't promise to send you a response. I get a lot of questions like this, but I'm happy to take a look with the web spam team just to double check that things are working out as they should. Why sometimes on one specific keyword I see maps in the search results and sometimes not? We try to recognize whether or not the user is looking for something that's geographically specific, and we'll sometimes show search results with the map, then. And that can differ, for example, depending on the device that you're using. Sometimes it makes more sense to show a map for mobile users. Something it makes less sense to show a map for a mobile user, for example. So that's something that really depends on the kind of the query and the kind of intent that we assume that the user has. One example, for example, is if you're looking for a local restaurant, and you search for a restaurant in your city name. Then that's a clear sign for us that you're looking for something specific maybe on a map. If you're next query is just pizza, for example, without a city name, then we could assume that you're actually still looking for something local. Whereas if you just search for pizza without previously doing anything else, then maybe you're looking for a pizza recipe, and you want a recipe instead of a local pizza place. So those are kind of the situations where sometimes you might see a map, sometimes you don't see a map. Is it possible to have two Google Analytics tracking codes on each page? Will this cause tracking issues? I can't speak for Google Analytics. I don't know for sure, but I think this is possible. I think I've done this in the past for some things. Mostly I did this accidentally and didn't realize I was doing it. But you probably want to double check with someone from the Analytics side, maybe in the Analytics help forum. I'm a simple blogger. I don't know anything technical. My competitor sends thousands of links to my blog. My blog is published by Penguin. In a case like that, if you're looking at the links in Webmaster Tools, for example, I might go ahead and submit a disavow file for those links there. In general, we do recognize these kinds of situations and handle them appropriately, but I'd also take a look at the other things that you're doing. So are you building links yourself? Are you kind of running multiple blogs and linking them between each other? Are you going to forums and posting things there? Are you commenting on other sites just to get a link back? All of these things are things that essentially are seen as natural links that could be problematic as well. And similarly, the content is also something that we focus on very strongly, so making sure that the content on your site is of the highest quality possible is also quite important. What I'd recommend doing in a situation like this is getting some feedback from peers, so going to maybe a local webmaster forum, maybe their local meetups for webmasters as well, maybe the Google+ groups that you feel comfortable posting in, and getting feedback from peers to see what they would see when they look at your website. If they see that perhaps there are quality issues that you could focus on if they see that there are things you're doing to promote your website which are outside of our webmaster guidelines, all of those things are definitely worth looking at with someone who's gone through these kind of processes before. So just because you see these links pointing at your site doesn't necessarily mean that they're the cause for anything specific with regards to your site in our search results, but if you do see these links, cleaning them up with a disavow file is something that doesn't take that much effort, so you could definitely do that if you wanted to do that.

GARY LEE: John, is it possible you can have algorithmic issues with links that are not listed in your Webmaster Tools, example, list of links?

JOHN MUELLER: Theoretically, it's possible. I think I'd see this mostly from the point of view, for example, if you have site-wide links from another site, then we only show a small sample of those site-wide links in Webmaster Tools. And of course, the rest of the links on on that site could be just the same. So if, for example, you have a site-wide advertisement on one website, and you don't have a nofollow there, then that's something we could pick up from an algorithmic point of view, and in Webmaster Tools, we'll just show a sample of links from that site. So that's something that wouldn't necessarily mean that you only have to put a nofollow on those individual URLs in Webmaster Tools, but rather clean up the site-wide issue completely so that it doesn't get taken into account at all.

GARY LEE: Sure, but more on a domain basis than regardless of-- do you tend to try and pretty much list every, not list all the good ones, but more listing the bad ones so that you can actually find them. Or is it completely random?

JOHN MUELLER: We try to focus on the more important ones, the ones that really would count more. And theoretically, if those more active ones are the ones that we find problematic, or kind of more problematic, whereas if they're individual ones that we don't list there that are problematic, then usually they're so small that they don't carry that much weight that they'd actually be really problematic. Theoretically, it could be that you have a ton of really small sites that are kind of problematic, and they kind of add up, but I think in most cases that's not going to happen.

GARY LEE: So somebody maybe being negatively SEO'd or whatever it might be going on, in a huge way they essentially, if they didn't realize it and then linked in [INAUDIBLE] you're saying that kind of scenario is very, very unlikely if not at all possible?

JOHN MUELLER: Yeah. I think that's very, very unlikely because usually you'd see that in the URLs that we show in Webmaster Tools. Those are the ones we primarily use for algorithms and for manual actions, and technically, or theoretically, it'd be possible that there are a lot of ones that are so small that we don't actually list them in Webmaster Tools, but there are just so many of them that they do add up to something quantifiable. But that's I think highly unlikely, I'd say.

GARY LEE: Yeah, OK. That's really good to know. Thank you.

JOHN MUELLER: All right. This is the same question as we had before. Does the position of the keyword in the URL help in ranking? So for example, if you have product name in the URL or not? We do take that into account slightly, but it's not something where you'd see a noticeable change if you had it or didn't have it in there. In the past, it's been that we show that in the search results. However, more and more we're showing bread crumbs in search results anyway, which are kind of the text information from there. So even if you just had IDs in the URL, then that wouldn't stop you from a ranking for those keywords. So this isn't something where I'd say you artificially need to put those keywords in there. If you have them in there, that's fine. If you don't have them in there, that's fine. You don't artificially need to change that around. We have a partner program that has been passing page rank through banners and links to our site which we've changed. This is all to be nofollow. Is this sufficient to recover from a manual penalty? Depending on what all your site has been doing, that might be sufficient, yeah. But you really need to take a look at the overall picture of your site, how it's interacting, how it's linking with other sites to see if that's really enough. What you could do is submit a reconsideration request detailing the changes you've made there, and you'll get a response back from the web spam team saying that, yes, this is good, we've dropped the manual action, or no, we still see issues, and maybe we'll show you some examples so that you can focus on those as well. The thing to keep in mind with reconsideration requests is sometimes they can take a few weeks to be processed, and it's not something where you'd want to iteratively send new ones all the time. So instead of like sending one today and one in two days and one in a week, maybe think about what you're trying to do, clean it up as much as possible, submit a reconsideration request, and make sure you've got everything covered that you might want to look at in the next couple days anyway. So clean up before you do the reconsideration request, and then give it some time to be reprocessed. Don't resubmit it again. All right. We're totally out of time and have a million questions left.

MALE SPEAKER: Do you mind if I ask one, since I'm on?

JOHN MUELLER: Sure, go for it.

MALE SPEAKER: Our site is under some kind of, I believe, algorithmic issue because we don't have a manual penalty. Is the fact that our site is now almost in all cases demoted to other results-- you know, click it, show all results-- is that a sign of any particular issue such as, I'm guessing, Panda because of a duplicate content issue? Or is it indicative of anything else in most cases?

JOHN MUELLER: That could be pretty much anything. Usually we show that kind of link at the bottom of the search results if we know there are other pages that are essentially the same as the ones we've already shown, so like you mentioned, duplicate content, that could be an issue there. And it's not necessarily a sign of any algorithmic or manual action that your site is being put in that place or that we're showing that at all for your search results, but usually just a sign that we have something that already matches that content in the search results before, and we that these additional results don't provide any extra value there. So that could be, for example, if you have an e-commerce site, if your descriptions match exactly what we've shown already, then we wouldn't need to show it again. But it's not a sign of any specific algorithm.

MALE SPEAKER: Right. I'll share this link in chat anyways just in case you've got a chance to look at it, and I think maybe I should speak to Gary afterwards, because it seems like when you take a snippet from one of our product pages, which is unique to our products, the scrapers are all, in every case, ranking above us even though if you look at that particular URL, the product from our site has sold thousands of times. It has 50, 60, 70 reviews on it. It has 50 people on Facebook liking it. And every other site is trying to be objective worse than ours because all they've done is taken our product. We have no manual penalty at all. I can't fathom how a human, if there was a manual review, which I would love, could look at that and say, OK, these sites ranking above you provide more value. Because you're pulling through our rich snippets, our reviews, everything, and still considering it worse than the other ones that have nothing, literally nothing, of value to add.

JOHN MUELLER: Yeah. I can take a look at specific site. We also have the forum to report scraper problems [INAUDIBLE].

MALE SPEAKER: Yeah, I've done that. I've been doing it for months. And I have emailed you two or three times after other Hangouts, John. I know you've got thousands of people emailing you. Perhaps Gary and I should form a club.

JOHN MUELLER: Unfortunately, you're not the only one.

[? MALE SPEAKER: ?] I know, but I try and be objective. I help others in the forums as well, as I'm sure other people do. But I look at it and say, OK, Google knows there's loads of reviews on here. Google knows it's our content. Google can see that, and yet it still chooses to promote other sites above us that clearly took it and published it later, but won't officially penalize us. So does that definitely say algorithm?

JOHN MUELLER: Yeah. I mean, one thing to keep in mind, especially when you're looking at individual cases there, is that we do tend to look at the site overall as well, so finding individual cases, sometimes it's a little bit easier than looking at the quality of the site overall. I'm not sure exactly which one your site is here. I can [INAUDIBLE] afterwards.

[? MALE SPEAKER: You only ?] have to look down to result six or seven. It's experience [? days. ?] It's the one with the reviews and everything.

JOHN MUELLER: OK, yeah. So for example, sometimes we see this in cases where a site-- I mean, I'll exclude your site from here, I've looked at it before-- but sometimes we see this in cases where a site primarily has content that's been copied from other sources and has individual articles, for example, that are unique and actually pretty good. And in those cases, we might see the site overall as being lower quality, and we'll have trouble recognizing that these individual pages are actually useful. So if you search for a snippet from something that's useful on that site, it'll look like, oh, this doesn't make any sense. But if you look at the site overall, we'd say, well, this kind of matches the overall experience on average from this website. I don't think that's the case with your website.

MALE SPEAKER: Every page is unique. And we're an e-commerce site, so they're all product, apart from a few categories, obviously. As people are saying, it is ripping the content off in the chat. We've been around for 10 years. Every single page is unique. I'm literally a bit like at wit's end as to say what is wrong with the site. People say, well, create unique content, build new links. The sort of links we've gained naturally in the last six months from "Fox News," "Huffington Post," all of these sites you literally can't buy, and yet there's still no change whatsoever. We don't do any [INAUDIBLE]. We don't do any SEO. We gave that game up a long time ago because we know it doesn't work. I mean, looking at it , can you can you say that that's--

JOHN MUELLER: I'll take a look at this page of yours, yeah.

MALE SPEAKER: Do you mind? Because that'd be great. And Gary [INAUDIBLE] exchange details.

JOHN MUELLER: OK. Great. I'll take another look at that one. All right. So with that, let's take a break here. I wish you guys a great weekend. Thank you all for joining. Thanks for all the big questions and discussions. It's been really interesting again, and I hope to see you guys in, I think, two weeks, or one and a half weeks? Something like that. Thanks. [INAUDIBLE] [? time. ?]

GARY LEE: Thanks, John. Excellent Hangout. I much appreciate it. Have a great weekend.

MALE SPEAKER: Thank you, John.

MALE SPEAKER: Have a great weekend. Bye.

JOHN MUELLER: Bye, everyone.

MALE SPEAKER: Bye. | Copyright 2019