 All right, welcome, everyone, to today's Webmaster Central Office Hours Hangout. My name is John Mueller. I am a Webmaster Trends Analyst here at Google in Switzerland. And part of what we do are these Office Hour Hangouts, where webmasters can join us and ask questions around the website and web search. And part of what we do are these little bit of echo somewhere. As always, if any of you want to get started with the first question, feel free to jump on in. John? Hi. All right, Michael. Good morning, how are you? Hi. Good afternoon for you. Recently, we've noticed that there was a little bit of a, not even a little bit, a few hour delay with stories getting into the index, even if we included the site dot, you know, WWW and the site's name, or the URL. We were noticing, I don't know if that, I mentioned it something to Danny a few days ago, I guess it kind of hit like maybe Saturday, say, do you know if that's been fixed? It's been taking kind of like a little bit of a while. And then when it hits, by the way, when it hits, it's fine. I have no complaints when it hits, but there seems to be this lag time. Yeah, I don't know what the current situation is there. I've seen a few reports of that. So you're probably not alone, but I don't know what the current situation there is. If that's something that's been resolved, or if there is still finalizing some things there. And one last question, not a major one, but I noticed when we link to something, whether it's a third party or our own stories, we tend to link on several of the words that sort of replicate the permalink, just because it's easier for the reader to know where they're going to be redirected to. But I noticed a lot of other places do it maybe on a word, like reports, or claims, or maybe even another outlet, just their name. Is there a best practice on that? We call that the anchor text of the link. So kind of the part of the text that is linked out. And we recommend having a kind of descriptive anchor text so that we can understand the context of the other page a little bit better. But it's something where we know lots of sites do that in weird ways. Sometimes you'll have something like, for more information about this and this, click here. And here is just the link, which makes it really hard for us to tell. But we try to look at the context of the link as well. So it's not terrible if we don't have a clear anchor text that leads there. It helps us to understand the other page a little bit better, but it's not like the absolute most critical factor. Yeah, now we try to do that because you don't want to direct someone the wrong way. But sometimes you have to kind of dance around to get the words that sort of match enough. So that's why I wanted to ask. OK, thank you. That makes sense. All right, let's see. Take that out of question as well. OK, go for it. Hey, so basically we have a website that we're in multiple different markets. So we're in Australia, US, and UK as our core. Our number one market out of that is Australia. But we're moving, we're taking TLDs. We're going from a dot travel to a dot com. Can we expect any type of, besides the standard traffic decrease that you would get from changing domains, can we expect any permanent decreases of visibility in our core market, Australia, from that move? Is there more of a neutral preference being given to the dot travel versus the dot com? So we would see both of those as generic top level domains. Meaning they could both be geo-targeted in any way that you want. So theoretically, if you do a clean migration from travel to dot com, and you really redirect one by one all of the URLs, then theoretically that should be about the same as before. So it shouldn't be that you'd see any significant drop in visibility. OK, also we don't have like a different, no, like we don't have a regionalized website either. All of our markets speak English. Is there, do you think there's any need to move to that type of regionalized folder structure yet? Totally depends on your audience. So that's something where I'd say if you don't need it, then don't add that extra complexity. As adding all of those different types of versions does make it a lot harder to maintain and monitor your site. And it makes it harder for us to show their site. Because you're essentially taking one strong page or one strong site, and you're splitting it up into multiple different sites. So we have to understand each of those sites separately. And they're essentially kind of competing against each other with regards to visibility. So I'd really only go down that direction when you really have a really strong business need where you say, well, I really need to give different content to users in this country compared to that country. Then that makes sense. But if you're just like, well, I'm a global business. Should I do it or not, then I'd avoid adding that extra complexity. Great. Thanks for your answer. Sure. Hi, John. Hi. Hi, I'm George. We started a editorial website dealing with the topic of electric mobility. And we took over a URL with our brand. After that, unfortunately, we found out that someone before us, maybe half a year, two years ago, used that URL for piracy content. And if we take a look in the Google Transparency Report, we can see that there has been a lot of trouble with that URL. Is there any way to get some kind of reset at Google? Because we do not have any connection to the former owner of this URL. And of course, we do not have any piracy content on our editorial offering. So what should we do? So this is something that settles down over time. I think I took a look at your site. You also posted a question, right? Yeah, correct. URL, yes. So I think your site is still fairly new. And that's something where it will settle down over time. But it's always a bit tricky if your domain has a lot of bad history associated with it. So that's something where I take a look at the old history there and think about, is this really problematic? Or is this a temporary phase that was involved with the site? And it's essentially just something fairly small that will just disappear on its own. Try to figure out if it's something that you need to take action on or where you just want to move forward and focus on your normal business and kind of over time, this old history will disappear on its own. OK, thanks. So you would not recommend to take a different URL, because that would be, to be honest, a very bad solution for us. Yeah, I would only do that if you really have a really kind of really, really bad history associated with that domain. So it's something like for the last 10 years, it was a piracy website. And it has lots of really bad or used to have lots of really bad content on it. And you just took over it. And you kind of need to have everything in perfect order now. Then that's something where you might say, well, I have to bite the bullet and pick a different domain that's a little bit cleaner with the history. But if it was really just a short period where it's, I don't know, a couple of weeks, a couple of months where someone bought the domain, used it for stupid reasons. And then that's something that usually will just bubble up, bubble down on its own. You can check in archive.org to see what the old content was and to kind of figure out, is this a longer term bad thing that was there or was it just a short term thing that will disappear on itself? Thanks. Sure. All right. So let's take a look at some of the questions that were submitted. As always, if you have questions or comment in between, feel free to jump on in as well. Let's see. Future requests, are there any plans for providing a download button in the new Search Console for the performance graphs? Currently, I can just download the queries, pages, countries, devices, and search appearance, but not click some pression, average position without using the API. I thought you can do that. Directly in the user interface in the new Search Console, I'll double check with the team and otherwise pass that on to them. Because that sounds like something that a lot of people would be doing. Does the alt image tag play a role in rankings for non-image search? Should we be putting descriptive comments on what the pictures are to help with general search results? So the text in the alt attribute for image tags is seen as a part of the text on the page. So that's something where it does help us a little bit to understand that page better. But for most cases, it'll be that we already have this text on the page somewhere anyway. So just by also having it in an alt attribute for an image tag wouldn't change anything. It does play a big role for image search, though. So if you have images that you want to have appear in Google Images, then make sure that you have a clean alt text there. Also, make sure that you watch out for the other aspects that relate to image search as well. Hangout to go, I asked about the above-the-fold algorithm and how it would be handled with mobile-first indexing. So basically, are you using the mobile version for the above-the-fold algorithm or not? We do try to use the mobile version as much as possible. So a lot of our algorithms are shifting over to using the mobile version of the page because that's something that most of our users see. So this is something where it definitely makes sense to focus on both the mobile and the desktop version when you're evaluating pages on your website and to think about it from that point of view as well. So don't just focus on what Google's algorithms do, but think about who your users are and what devices they're using and what they would see. Let's see. Searching for a brand name brings up the Knowledge Graph. However, the logo gets taken from a different source. Then our organizational markup points out. Image size and implementation is, as the guidelines state, any idea what we can do. I took a quick look at this. And looking at the page, it seems that we're pulling out the same logo that you're showing on the page and also matching the same logo that's shown on the Wikipedia page. So I could imagine that our algorithms are a little bit confused there because for the logo markup that you have on the page, you have a completely different logo listed there. So that's something where I imagine our algorithms are getting a bit conflicting information there. And the more you can make things align for us, the more likely you will be able to pick up what you have in the mark. All our sites currently serve a standalone interstitial for country language selection based on the IP detection. Users have to dismiss it in order to keep browsing sites. Is this OK? Or am I at risk of Google not being able to see the content behind the interstitial? Yes, that's a possibility that we wouldn't be able to see the content behind the interstitial. So our recommendation for these kind of setups is to use a banner instead of an interstitial. So if you can recognize that the user is coming from a different location than your page's target, show a banner on top so that the user can select the other location and go to that version instead, but that the user still has the option of seeing the page in the original format that they clicked on. So sometimes, for example, a user might be in a different country, but they're actually trying to look at the page in their own language or for their previous country as well. So that's something that helps users there. And with regards to Googlebot, Googlebot primarily crawls from the US. So if you show all users in the US an interstitial instead of, for example, the French content, then we would never be able to reach the French content. We would only see the interstitial. We would not be able to crawl and index the French version. Whereas if you use a banner on top, then we would be able to see the banner, but we'd still be able to see the rest of the content on that page as well. So definitely aim for something less intrusive, like a banner, rather than a complete interstitial. How would Google react to if it sees the H2 above the fold and an H1 below the fold? Also with HTML5, you can have as many H1s as you want. How about different H2s? And can you have several of those as well? We're definitely very flexible with regards to our understanding of headings on a page. So we use headings to better understand the context of the content that you have there. And if you have an H1 on top, or if you have an H1 on the bottom and you have a little bit of context there, then that's perfectly fine for us. We don't have a limit with regards to the number of headings that you can have on the page. I would really focus on what makes sense for your page, for your users, and look at it that way, rather than to think about how Googlebot might be limited from there. I know a lot of people like to say that you should only have one heading on a page, or one primary heading on the page. And sometimes I think that's a good practice. But Googlebot is used to seeing a lot of different kinds of pages. And some pages have one heading, some have no headings, some have multiple headings on the same page. And we have to be able to deal with all of that. So I wouldn't worry too much about that. In the past, you explained that Googlebot or Google is not researching author backgrounds, expertise, et cetera. Can you say the same thing for site reputation and Better Business Bureau scores? For example, some believe that BBB ratings and reviews are used algorithmically with the latest core updates. That doesn't make sense, since the BBB is only for the US, Mexico, and Canada. I can't imagine that Google would use a single source like that algorithmically when its algorithms are mainly global in nature. I would venture to guess that you are correct, that we wouldn't use something like a BBB score for something like this. As far as I know, that's certainly the case. There are various kind of issues with regards to some of these sources of information about a business, about a website. And we need to make sure that we're really reflecting what we think is actually relevant for users rather than blindly relying on some third party's ratings. If we put the text from customer reviews on our website, can we put the same review text on more than one page? So you can put the same review text on more than one page, of course. But when it comes to marking up reviews with structured data so that they're shown in search, those reviews must reflect that kind of primary topic of that page. So if you have a product that you're selling, those reviews must be about that product, not about other products on your website. And it should be the case that those reviews accurately reflect the overall possibility of users that have been leaving reviews for your site. So it shouldn't be a testimonial. It shouldn't be a hand-picked selection of reviews. It shouldn't be copied reviews from other websites that you're reusing. It should really be something that users are able to leave on your site directly. Hi, John, regarding the reviews, what happens if I have multiple reviews and the review will change every single day? Based on the guidelines of the structured data, should I mark, let's say I got 10 reviews on the pages, but those 10 reviews will update every single day? Should I mark up those 10 reviews? How do you mean they change every day? Because there's a new review comes up for that product. Like, people purchase this product every single day, then the review will be fresh, and then the legacy one will go to the second page. OK, so it's not that you're hand-picking different reviews every day. I think that's perfectly fine. If these are just kind of new content that's coming available on that page, and you're moving the other reviews to maybe like a review archive page, if you will, where you have a collection of all of the reviews, then that's perfectly fine. So Google can update it once the review is being updated as well? Exactly, yeah. Hey, John. Yeah. I've got a question. So we're into some of the first indexing. So we have a website, or several. And from the home page, you can click into a category. And on the category page, it lists, OK, here are the sub categories, and here are a set of products. And you can either check out the existing products or click further into subcategories. On the mobile site, we have the ability to click into a category that you don't see any products until you click into the deepest category level. And sometimes that might be three, four, or even six clicks deeper. How big an issue is that in the difference? I imagine what we'll see is those product pages not performing as well. It's possible, yeah. So with mobile first indexing, we would only use the mobile version for crawling and indexing your site. So if on the mobile version, those pages are linked in a suboptimal way, then that's what we'll have to use as a basis for crawling and indexing the rest of the site. So that's something to kind of think about there. What I'd recommend doing in a case like that is maybe checking with one of the third party crawler tools that are out there to see how your website is crawled in the desktop version and how it's crawled in the mobile version and to try to look at the results and think about is this really bad or is this just slightly different? It's like this one crawls like this and the other one crawls slightly differently. Ultimately, we reach the same pages are kind of linked in some other way. Maybe that's OK as well. And a follow-up question somewhat related again to mobile first indexing, impagination. So on a desktop site, as you look at a set of products, you have the ability to bounce from page one and you can go five clicks off of that. And on mobile, we shrink that. What's the best practice there? Because currently, they're showing just a next button to bounce to the next page as opposed to a larger set. And is it appropriate in a responsive situation to show fewer to the user but still have those links embedded? I think there is also no absolute answer for that case. It's really just the case that we would crawl and index the mobile version there. So if the mobile pagination is different, that on its own is not a problem. But we'll only use the mobile version for the pagination. So if the mobile pagination doesn't work at all, then that would be a problem. If it's just slightly different, then that's not a problem on its own. But if it negatively affects discovery of your content or making it visible on the page, then that wouldn't be visible in the search results. So it wouldn't be considered a problem to have a responsive pagination that shrinks the number of visible links available on the page. That would work if it were. Okay, great, thank you. Sure. Regarding the pagination, I got a question really to do that is if I have pages, we'll bring to second page and third page. Of the canonical, should I also set the canonical as the second pages and then? Yes, okay. Yeah, and then we have to use the next for the pagination as well. Exactly, yeah, okay. So then Google will crawl the second and third, okay. So I don't need to, all the pagination pages are chemical back to the main ones. Exactly, so if you set the canonical back to the main one, there are two things that could happen. One is that we look at the pages and say these are different pages. We'll ignore the canonical, which doesn't help you. Or what could happen is that we follow your rel canonical and we only index the main, the first page. So if there's a link on the second page that is critical for finding new content on your website, then if we only focus on the first page, we'll never see that link on the second page. I see, thanks. Sure. All right, the file of spam report tool is woefully inadequate. Why is there not a better way to report spam? We have competitors that are top in the market, but they're using verifiable fake testimonials, aren't actually at the address they claim and have tons of fake reviews on third party sites and they're buying links. So how do we report all of that to Google? That's something that sounds like you have quite a collection of stuff there. If you want to, you can send that directly to me. On Google Plus is probably the easiest way to do that. You can send a private message with the details there. In general, with web spam reports, what you can also do is link to a document with more details in there. So in particular, if, for example, you're finding a bunch of paid links that are pointing to a website and you'd like to report that to Google, then put a spreadsheet online on Google Drive that people can access, that when we see your spam report, we can go to that spreadsheet and get those information. The important thing with the spam reports is that there is a limit for the text field that is there, and that's primarily so that the spam team can take a look at what you're reporting and fairly quickly figure out in which bucket that belongs. So if you write us a really long novel about all of these different things that are happening, and it's really hard for the spam team to understand what you're trying to say, what the quintessence of your spam report is, then that can result in the website saying, okay, we need to kind of shelve this for the moment and look at it when we have a bit more time. So kind of shortened to the point, that really helps when it comes to link issues, including a link to a spreadsheet, that kind of helps. And if it's something where you really have a lot of comprehensive information about a bigger issue that you'd like to forward on to Google, you're welcome to send out to me or to one of the other people that are working kind of with regards to web search externally. I'm looking to improve our SEO for international markets and we're currently looking into options on the URL structure. We have nailed this down to two options, subdomain or subdirectory. Oh my God, these subdomain, subdirectory questions. Okay, let's see, how does it continue? We have a strong brand for our fans and decent SEO ranking, but we'd like to improve that for international. What's the most efficient URL structure in order to improve rankings for international markets? So I assume you're looking, you have a generic top level domain like a .com website and you're looking at en.example.com or example.com slash en maybe for English or maybe you have different country codes that you're using there. And from our point of view, both of those options are perfectly fine. It's something where I would look at more into kind of the details of how you want to maintain that in the long run are these essentially the same site on the same server and it makes sense for you to keep it all on the same host name but maybe subdirectories make it a lot easier for you. They certainly make it easier to maintain things like search console information. You just have one site that you need to verify. You can use the subdirectories to set the geo targeting but you have all of the tracking information in one site in search console. So that's kind of useful. Also all of the settings that are host wide that are server wide, they're in one place. So you can set the maximum crawl speed. You have one robot sex file. There are kind of fewer things that can break there. So from that point of view, I think that generally makes sense. However, if you really need to split this up for whatever organization own means into different servers or different hosts, then using subdirect subdomains is also an option. So both of those are options that I would say would work in a case like this. Generally, I'd recommend using subdirectories here just to keep things a little bit easier. I have a question. It's my question, so thank you for answering. My question is related to subdomains. I have read through articles that if I would use the subdomain approach. So let's say I have my website is website.com and I have a specific ranking on sale. I've been working a lot of this and stuff like this. I also start using subdomains. Let's say I'm pushing for German market and we'll be like de.website.com and articles are saying that I would lose my ranking from my generic website when I start using the subdomains. So what you would certainly see is once you start splitting things up, you would see probably a drop in rankings because we would take, so for example, you take one page that you have currently for your global market, you split that up into one page for Germany and one page for non-German, let's say. Then essentially you're taking one piece of content, splitting it up onto two URLs, and that means we have to rank those pieces of content individually. So in an ideal situation, that's maybe a 50-50 split, which means that these pages on their own individually kind of don't have as much strength anymore on their own. So if you take that to a subdomain or if you take that to a subdirectory approach, then in both of those cases, you also see that drop where we say, well, you have one piece of content, now you have two pieces of content, and they have to stand on their own. So we focus on them individually. With subdomains, what you might also see is that initially, when we start seeing this, we might not be sure if this is actually part of the same website or if this is meant to be a different website. So sometimes subdomains are used for completely separate purposes, maybe separate brands, maybe even separate companies. So initially, what might happen is that we look at that and say, oh, well, there is a completely separate here on de.example.com. And it'll take us a little bit of time to understand, well, actually, these subdomains are all a part of the same website. So that might also kind of flow into that situation where you'd see maybe an additional drop in rankings for that kind of case. OK, so I might assume that if I would take the subdomain approach, I might get some drops in my rankings. Because I won't use the ranking from my example.com main website. Probably. So at least initially, in the long run, that should settle down. But any time you split things up, you'll see this kind of drop in ranking. So even if you go with a sub-directory approach, you're very likely to see the situation where algorithms suddenly see two pages instead of one. And they have to reevaluate how these pages individually should be shown in the search results. So that's also a good reason to limit the number of different country versions that you create. And don't just blindly go out and say, oh, there are, I don't know, hundreds of countries. Therefore, I'll create hundreds of different subdomains or subdirectories for each one of these individually. Because then you're really splitting things up into all of these different charts. So really only focus on the aspects that you absolutely need for your business. And then think about what is the long-term approach that you want to take here? Do you need to have a separate subdomain? Or can you keep everything on the same host? If you can keep everything on the same host, then I would recommend doing that just because it's also much easier for maintenance long-term. Thank you. Sure. All right. Can John go three? Oh, go for it. Sorry, John. I have a similar question to that. So I thought maybe I'd jump in if you don't mind. OK. I'm running into a situation a couple of times where I have sites that are ranking well, but they're not ranking for what the client wants them to rank for. So like they have a ring site that sells rings. They also, for some reason, have a blog. And they're blogging about rings. And the blog is ranking great. Everyone loves their articles about the 16 millimeter diameter gold ring or whatever it is. But they can't sell a ring to save their life. And so what do you recommend in that scenario? I thought we could either separate out the blog into a subdomain and try and get Google to rethink about the e-commerce part of the site. Or I even thought about drastically moving the blog off to a totally different domain name altogether and trying to get Google to rethink about the e-commerce site. What do you recommend in that scenario? I don't have any kind of standard solution that would help there. But I would be tempted to try to keep both of them together. I don't think there should be any problem with that. So I mean, in the worst case, the same site ranks. But it's not that I don't think we would rank the e-commerce site better just because there isn't a blog associated with it. Kind of this collection of multiple things on the same website that's very common. That's not something that I would try to avoid completely. And so when Google is preferring a blog page for no-do queries and when we get down to do queries, we want Google to select the sales pages. That seems to be exactly where Google is getting confused. And so the only way I could think of was 301ing the no-do to the do or separating the sites out and making Google completely rethink the pages. But you think they can all stay on the same site. And we just have to improve the quality then of the do pages so that Google says, OK, these guys also sell ranks. Great, let's rank those. Probably, yeah. So usually that's really just a sign that we think these blog pages are more relevant for these particular queries, which might be the, I don't know, the situation and what people are searching for there. But that could certainly be the case. But yeah, I don't think there is kind of this one answer if it's all for that kind of situation. And what I would also take into account is kind of the other content on the blog. Is it all themed for the same thing? Or is there some completely random content on the blog? And if it's kind of partially overlapping but not completely overlapping, I think it might make more sense to split the blog up to move it into a separate part from the site. But if it's essentially similar content to what you have on the rest of the site, then I think leaving them together should be fine. I see. So if they also had a section on bracelets, but they didn't sell bracelets? I'm thinking more like they had a section on interior decoration. And they were kind of focusing on jewelry. That's kind of a mismatch. Yeah, here's our articles on rings. And also, did you know you can paint your walls like this? That's kind of the thing where I mean, sometimes these blogs do go into kind of personal preferences and look at this cool room I created. But it can make it a little bit trickier for us. So I try to find one main focus for the site. I see. OK, thanks, John. All right. Yeah, John, actually I wanted to know the links browser. Have you heard of the links browser? So that browser is showcasing the website. So I have read somewhere that the browser shows us actually and the search engine looks at the website. So is it true if my website is like links is looking at my website and the search engine is looking at my website in the same way? Is it like I mean? So the links browser is a text-only browser that I don't think supports anything around JavaScript or CSS, any of the modern HTML features. So it would be something that reflects what a very simple search engine would look at. But I don't think it would reflect what a modern search engine looks like. Yeah, I just wanted to know if, in that case, how a search engine will index images also. So that's why I just wanted to know. Yeah, I would use the tools like fetch and render in Search Console to get a better view of what Google, for example, would index. The links browser is really very simplified. So I think it's good for debugging some very basic issues. But for the most part, I would focus on the more advanced tools out there. Thank you. Thank you so much. All right, can 503 server responses have a negative impact on crawling and or rankings? So 503 is a temporary server error that, when we see that, we'll generally back up, back off from crawling a little bit. So we'll slow down crawling. And we'll come back to those URLs a little bit later. If we see a 503 error persistently returned, then we'll assume that it's a permanent error, not a temporary one. And we'll assume that these URLs that have been returning 503 are actually no longer available on your website. So we'll drop them out of our search results. So if your website is temporarily down and you server 503, that's perfect. If your website is actually up and you use 503 instead of a normal server response code, then that would be a bad thing. Why is the publisher Carousel showing articles of two months ago for our site? What can we do? We published many articles about this topic while in the last two months. And it happens in English and different language versions. I don't know. I'll take a look at that with the team here to see if there is something more specific that we can say there. In general, these carousels that we show in the Knowledge Graph on the side are things that we algorithmically generate based on the website itself. It's not something that would be reflecting the normal website's ranking otherwise. So it's really only people are looking for your website specifically by name. We show that in the sidebar in the Knowledge Graph. And we'll pick some of the articles that we found there and highlight those there. But of course, other articles from your website will be shown in the normal search results as well. So that's something where there are different ways that you can be visible in those kind of search results. In faceted e-commerce sites with multiple combined categories, for example, t-shirts and dresses, how do you recommend to split the category in two? Google doesn't have an inverted canonical tag to kind of say, split it like this. What can you do there? So my general recommendation is to keep one URL and just create a second one. And obviously, the second one you can't redirect to because if you redirect, then the original URL is gone. But rather, just update the internal linking within your website to point to that new URL that you have there as well. If you can't keep the old one, for example, if the old URL, in this case, is your site.com slash t-shirts and dresses, then maybe you want just one for t-shirts and one for dresses, then I would redirect the old URL to one of the new ones, either for t-shirts or dresses. And otherwise, just link the new URLs normally within your website's structure. So there is no specific markup that you can do to say this URL has been split into two halves, or this part of the content has gone here and this part has gone here. It's really essentially just a new page that you create on your website. And it's associated with the old page, maybe with a redirect. Maybe it's just a completely new page. Is the structured data report going to be moved across to the new Search Console from the old Search Console? I don't know. I know that the Search Console team is looking into various of these reports and moving them over or regenerating them as we go along. I don't know what the specific plans are with regards to that specific report. And generally, we wouldn't be pre-announcing those specific changes ahead of time, but rather announcing them when they're available or when they're available for testing. So I'm not quite sure how that one would map over. It's always a bit tricky with the structured data reports because we have so many different variations of structured data testing tools, the rich results testing tool. Then there's a rich cards report and rich results report and a structured data report. And I think in an ideal world, it would be nice to have one combined report to handle all of these different situations and maybe one combined tool to handle all of these different variations as well. I don't know how much of that we'll be able to combine in Search Console because obviously, when you combine things, you also make it a little bit more complicated because suddenly, instead of one simple way to test the URL, you have different variations within the same tool. So we'll see how that works out, but I know the team is continuously working on these topics as well. Is there a way to tell Google not to crawl and index certain parts of a page? For example, we might want to have some user reviews on multiple pages because we think this is helpful for the user, but we don't want Google to penalize us for duplicate content. Can we tell Google to index our pages but skip a few paragraphs from them? There is no simple way to hide some content on a page from Google. They're tricky things that you can do with regards to using an iframe or using JavaScript to pull in content that's otherwise blocked by robots text. But in general, we prefer being able to see the page as a user would be able to see the page. So that's something that you can't specifically do there. And what is perhaps more important here, you don't really need to do that. So it's not the case that we would penalize a website for duplicate content if a part of the text on a page is the same as elsewhere on the website. So that's not something that you need to kind of find a technical solution for because that's very common on the web. We see a lot of content that's shared across a website. Sometimes it's just a footer. Sometimes it's a giant legal disclaimer that's on every page. Sometimes there are different kind of product descriptions that are shared across different variations of a product. All of that is perfectly fine. And it's not a reason for us to demota a website or to give it a web spam penalty or anything like that. Excuse me, I have a question. Sure. Yeah, I could just jump in. Go for it. OK, I appreciate it. So I'm new to the whole Google Hangout and everything. So I have a website, a mental health website that I've had for a year now. And I've been writing a lot of content for it, pretty much like five articles a week. And I've just recently gotten a lot of really high quality backlinks from really high authoritative websites. And we're really about last month is when I started reaching out to do that. My question is, so this website is a year old. Like, how many years or when can I expect significant search traffic? There is no specific limitation there. So it's not the case that there's any kind of sandbox effect where we would hold the website back and say, oh, this website needs to kind of maintain its presence for a year or two years before we can start showing it in search. It's really more a matter of us having the signals that are available for that website and being able to show it. So that can happen now. That can happen when you start publishing new content. That can happen over time. Or it might happen that it never actually happens. So maybe you're working on this website for multiple years, but it's still not something that we would consider as being relevant for those particular queries. So maybe it would never see that much visibility in search. But at least there's no artificial limit that's holding this website back and saying, oh, you need to have, like, minimum this amount of time on the internet before we can start to bring your website appropriately in search. OK, so there's just a million different factors. We use, so we say we use over 200 factors for crawling. So a million might be a little bit too much, but a lot of it. And especially when you're looking at competitive areas, then that's something where some sites might have been there for a really long time. They've built up all of the kind of representation for presence online for years and years. And going in there and disrupting that is something that really helps to go in there and say, well, I have this new spin on this topic that is really novel and not something that other sites have at all. You kind of have a chance to be a little bit different from this. And that makes it possible to kind of be kind of jumping in there as well. It doesn't guarantee it. Of course, maybe your site is just very different and different in the battery. So those are always different options there. Thank you. Something I have noticed, though. I've noticed a lot of really just simple websites, kind of spammy-looking, the content. A lot of the content is just the same, and it's just slightly reworded for the keywords. And they're ranking at the top in Google. And they're super old. They're probably websites that have been in Google for 10 years or so. And I'm just wondering, how are these websites at the top in Google when they're just so low quality? I don't know if they've just been grandfathered in after all the Google updates and everything. I don't know. I don't know. If you find that they're spammy, I would definitely submit a spam report. But the WebSarm team can take a look. If you're just not sure about the quality, you can also send me a note on Google+. And I can pass it on to kind of the quality team to take a look and see are our algorithms doing the right thing. OK, thanks. I'm sure other people have questions, but I appreciate it. Thank you. Sure. All right. More questions from any of you. Can I ask you a quick question? Sorry, may I ask you a quick question about entities? OK, I'll try. Is it hummingbird that is the algorithm there that deals with entities for you guys? I think so. I think that's kind of when we started talking about entities in search, yeah. That was June 2013, if I recall. I don't know. Long time ago. We memorize these things, John. We watch these things. It's a fun question. I just want to ask what kind of websites are good candidates for Google, for hummingbird in this case, to find entities and to kind of categorize entities? Like we know we use Wikipedia. What other kind of websites do you think are good candidates there? LinkedIn, Google My Business pages, et cetera, et cetera? I think we use a lot of different sources. And what helps us there is also structured data on these pages. So the better we can pull out the information in a machine-readable way, the more likely we can use that to understand kind of the topics and the entities that this page is talking about. I see. And so some little blog somewhere would not be a good candidate to parse entities. Sure. I mean, we can pick up that content, too. So if we have structured data there and it's in a way that we can say, well, this is trustworthy, this is useful information there, then that's something we should be able to pick up there, too. But the data really has to be structured, as you said, and specifically using microdata format, or? I mean, it's easiest if we have it in structured data because then it's in a machine-readable way. Right, right, right. OK, so machine-readable really is the key. I guess a well-formatted HTML page would suffice in many cases. That would also help us, yeah. OK, great. Fantastic, thanks. Cool. All right, Mihai, you had a question. Hi, Joe. Hope you was fine, happy to see you back. So a quick question about regarding a website that we're working with. They basically have software as a service product. So they have a website that presents the features of the product, how can you use it, a lot of content regarding different kind of topics related to the product. And now they've also built a sort of a tool that kind of indexes a huge amount of pages from the web regarding things, oh, how can I put it? Messages, emails, anything that's regarding issues surrounding certain IT products, like from GitHub or from Stack Overflow or a lot of that stuff. So they made a search engine where you can actually search for all of the data. So obviously, and they put it just to show you. So that is the website, and the second URL is the tool. And so that tool creates millions and millions of pages where before it was just a simple website with 100 or so pages with content and product features and everything else. Is there any way that can just creating those minion-plus pages? Is there any way that can affect us? And I'm less interested in rankings, but more in terms of crawling and indexing. So would it be a better idea to use a subdomain for that with the Google crawler kind of allocate a separate budget so it wouldn't affect the pages from the main website? So to me, this sounds like it's scraping content from the rest of the web and spinning it together. So that kind of goes into auto-generated content, low-quality content. So this is something I prefer to see blocked by robots text in general. And if you have it blocked by robots text, then it doesn't really matter where you have it on your website. So that's a lot of that content isn't actually indexable by search engines. So the way they source it, a lot of it isn't found on any other website. This is why they created this search engine because you cannot actually find most of that information on Google. I don't know. So compared to some of the other sites that I've seen, do something similar. For the most part, it's really low-quality content. And when you're automatically generating this content, that's something that I could imagine our algorithms looking at that and saying, oh, this is really low-quality stuff. We shouldn't be crawling and indexing this content. And they might back off from crawling the content. They might back off from crawling and indexing the rest of the website as well. So that's something where you really, really need to make sure that you're providing significant, high-quality, unique content here, not just like a new spin on the same topic. So things like we see them a lot, eBase grapers or image search grapers that take information from LinkedIn, from various search results, maybe other search engines, maybe internal search results from other sites, and compile that on a page of its own. And that's something that's really low-quality, kind of specific type stuff. So that's, yeah. I know what you're referring to. I don't think it's that I have to look, obviously, more into it to make sure that it's not. It's apparently a very highly used tool. A lot of people seem to use it. I'm not very familiar with it yet. But let's say it's no problem from a quality point of view. Would it still make sense, would my questions make sense, whether we should move it to a subdomain so it doesn't affect the crawling, so that Googlebot doesn't really just focus and get lost in that sea of URLs? I wouldn't treat that any differently than if you're adding other content to your website. So that's something where, for the most part, I would just keep it on the normal host name and just see like you're putting an archive of material on your website. And it's available now, and it can be crawled and indexed. Even if that content isn't really very much related to the rest of the website? I don't know. You're starting to sound like that spamming paper guy again, or it's like, I have enough content. I don't know why I have it. I have a big situation where I'd say, well, how do you know? It's really high quality content if you're like, sure, you can make it to anything else on the website. So I would treat it the same as like you're taking a collection of content that you have, and you're putting it online on your site. And that could be like, if you're a newspaper, it could be an archive of old things. It could be some other catalog that you have available that suddenly you'd like to make available online. Kind of see it as that rather than try to see it as something unique based on this specific piece of content. But I'd really be critical with regards to the quality of the content and make sure that you're not just creating this universe of spam that will essentially be a big hassle to clean up. OK, I'm very sure it's not spam. I just have to really see if it's worth letting search engine index it, whether it does get a lot of traffic, basically double the traffic of the website as soon as we launched it. All right, I need to hand out. Thank you all for joining, and hopefully I'll see you again in one of the next Office Hour Hangouts. Bye, everyone. Have a good day. Bye, John.