 for technical problems. All right, welcome, everyone, to today's Webmaster Central Office Hours Hangouts. My name is John Mueller. I am a Webmaster Trends Analyst at Google in Switzerland. And part of what we do are these Office Hour Hangouts, where people can join in and ask their questions all around their websites and web search. Tons of stuff was submitted already. So I'll see if I can get some answers into things that we don't get to in the comments as well. But if any of you want to get started, you're welcome to jump in with cross-question. I can try one, I guess. All right, go for it. So more of a theoretical one. Let's say you have a store or maybe a site running a classified ad, something that's fairly big. It's doing very well. And it kind of targets a lot of topics. So if you have a marketplace or maybe a classified ad, you might have pages regarding jobs, pages regarding selling fashion items and things like that. And let's say you're doing fairly well for most of the topics you're targeting. But most of the websites that rank in the first positions are dedicated to that specific topic. So maybe it's hard to beat a fashion store with your fashion section of your marketplace when you have all these other sessions. And what would happen if that website decides, well, let's just create a new website dedicated to just the fashion section that we have. And we like to run two websites in Pell or something like that. And that would mean the content is kind of the same. So one product uploaded somewhere, it shows up on the other website as well. How does Google treat this other than just two websites that have similar content? Does Google like the same business running those two websites? Does that play any role in how Google decides what to rank, what to index, and things like that? I don't think we would do anything special in a case like that. We would essentially try to treat it as two websites with regards to whether or not that makes sense. I think that's a bit of a different question because sometimes it makes sense to have one really strong website that includes a lot of different topics. Sometimes it makes sense to focus a little bit more. And finally, that balance is more, I'd say, a marketing and kind of strategic problem rather than SEO problem. Because what won't happen is that if you're ranking at number three, then suddenly you're ranking at number two and three. I don't think that would just happen automatically. But rather, usually you'd probably, I don't know, try to link to your other website or maybe redirect some pages to that new website that you have and that dilutes a little bit of the value of that big, strong website that you have and creates a little bit more value and that more concentrated website then. And whether or not that works well for search or that works well for users, that's really hard to say ahead of time. With regards to two websites like that, I don't see a problem. If you're going in the direction of I'll create 10 or 20 websites like that, then that's something where I could see the kind of quality algorithms kicking in and saying, oh, this looks a bit like doorway sites. We need to be careful here. But if it's just two websites, if you decide to go to three websites at some point, I don't really see a big problem there. I'm asking since over here most of the classified websites, they've also built websites dedicated to either the jobs market or the real estate market. So anything somebody posts an ad on a classified ad on the jobs section of the main classified site, it automatically appears on the job specific site as well. So it's like duplicating the section of your site and creating a whole other website for it. And I mean, they both seem to rank well for keywords related, let's say, to the job market and things like that. So I was wondering whether Google has any issues with that or might be any issues that somebody should take into account. I don't think we would have any issues by default. I think it's really more a kind of a strategic question. Do you want to go down that direction or not? There's a lot of work involved with splitting things off and kind of duplicating things, finding the right balance between how much you duplicate, how much you don't. But maybe that trade-off is overall positive for that site, for that situation. I don't know, it's hard to say. If it's duplicates, let's say the same pages, the same content, classified ads show up on both websites. Is there a situation where we might decide that, well, we've already seen this content, and it's exactly the same on this other website, doesn't make much sense to show both websites? Yes, a really simplified way to look at it is if the snippet would be exactly the same in the search results, then it makes sense for us to filter one of those out. And that depends a little bit on the query. If people are searching for something that is kind of more specific to one of these websites or not, that's usually kind of the simplified approach to look at it. If you think that the snippet will look exactly the same, then we'll filter one of them out. If the snippet would be different, sorry, then we would show them both. And if Google does decide to filter one of the sites out, is there any way to see that in a search console? I don't think so. I mean, other than you just don't have any impressions for that query or whatever. If you have two websites and you look at the query statistics, you can probably see from the impressions, like, does it match, does it not match? But it would show up at the coverage section. So the coverage section would still show the websites as being indexed? Yeah. Yeah. Yeah, they would be indexed normally. And if someone searches for, I don't know, a keyword from the classified ad plus something from the location or something from the title, then we would just show that one. Or maybe if they search for something slightly more different to the other one, then we would show the other side. So indexing would be fine. It's really just showing in the search results that would be different. With regards to indexing, what would be problematic is if the exact same page were used on both of these sites. If you use the same layout, the same HTML page, if that was exactly the same on both of these sites, then we would see that as duplicates because it's exactly the same thing. Yeah. Cool, thanks. Sure. All right, any other questions before we jump into the submitted ones? Hi, John. Hi. I have a question about a singular and plural form of keyword. So one of our clients, that keyword is garden shed Sydney and garden shed Sydney. Now, for garden shed Sydney, the category page, garden shed category page ranks on Google. But for the singular form, garden shed Sydney, one of the blog posts is ranking on Google instead of the category page. Why is it different? Both keywords are the same, just singular and plural. So why is that difference? So it's hard to say without looking into the exact details, but we would see those queries as being different. I think kind of the first step there. And when we see them as being slightly different, then we might think that one or the other of these pages makes more sense to show. So usually with singular and plural, we do recognize that they're synonyms, more or less. But we also recognize that maybe there is something kind of unique to one of them or to the other one, such as if you're looking for a plural, maybe you're looking more for a list or a comparison page or maybe a category page of different kinds of these items. So that's something where our systems try to take that into account. And it can result in slightly different results being shown for one or the other. It's a bit tricky when you're in that situation, you're like, oh, but I want my other page to rank instead of this one. And you don't want to remove the page that you currently have ranking. That's something where you can't really force that other than to tweak things subtly. That you kind of make sure that the right words, the right phrasing is on these pages, that you link them internally, properly. But that's sometimes kind of tricky. It's also worth keeping in mind just because when you take a step back that these words or these queries sound very similar and they seem very much the same, it might well be that users do treat them as different queries and do expect different kinds of results. So before just jumping in and saying, oh, I need to have the same page rank for both of these, maybe check with some other people to see does it make sense to change this or is this something where it's actually not that bad? Or another thing you could do is the page that's currently ranking, put some kind of a call to action on it and say, hey, if you're looking for this, also check out this other page. And another question, John. So we have a client who they are a startup management company. They have a business in five states in Australia. For each state, they have different websites and they have one corporate website. Now, their corporate website has a block, but the other website does not have any block. Now, the corporate website, the blog of the corporate website, it gives link to the state website. Will there be any problem? So what's the last part? OK, so the blog of the corporate site gives link to the state website. Will that be a problem? OK, so the blog just links to one of them or to all of them? Or? It depends. They don't give link every time. When it's necessary, they just give link to one of the state site if it is given to that site. I don't see a problem offhand with that. I think this kind of setup that you have one big corporate site for kind of the overgroup and then the individual locations or the individual some companies, a separate websites, and then kind of link together, I think that's pretty common. That's not something I would consider problematic. Thank you, John. Sure. All right, let me run through some of the submitted questions. And as always, if you have any comments or more questions along the way, feel free to jump in. And towards the end, hopefully, we'll have some time for other questions from you all. Our website and our content is quite usually getting copied by someone else. They basically just create a new domain and copy the whole website on it. Is there something, apart from disavowing these sites that we can do, can we submit these sites as duplicates so Google won't harm our results and sites as a result? I think this is always kind of tricky. The general approach that I would try to take here is to see if there's something, from a legal point of view, that you can do to essentially tackle the problem at source. Because if the website is no longer up, if it's no longer being indexed, then that's essentially a solution for search as well. So I would double check with whoever you work with on a legal side with regards to things like DMCA and copyright to see if there's something that you can do there. With regards to the location of the website, if they're hosted in other countries, all of those things, that might be an issue with regards to the full legal process, but with regards to how Google processes these reports. Generally, that's less of an issue. So that would be the first step that I would take there. And alternately, what you could do is, of course, submit these with the web spam report. With the web spam report, it's important to keep in mind that this is not something that we go through one-on-one and kind of manually review everything and throw all of these sites out. For a large part, these are used essentially to train our algorithms to improve over time. So just because you submit something to the web spam report, I wouldn't expect that to kind of be removed automatically after a couple of days or something like that. But those are kind of the approaches that I would take there. With regards to disavowing those sites, if they're linking to your website and you really don't want that link to be taken into account, then using the disavow tool is fine. If they don't link to your website, if they're just copying your website, then there is nothing really there to disavow. Hi, John. Can I have a question to that? Sure. So there might not be a possibility to take those legal steps you're talking about. But when it comes to Google, is there some possibility Google will harm our website? Or can Google see that our website has history and is much bigger, has much more authority than the copies? So therefore, it's basically OK? I don't know about basically OK, because you submitted this question. It sounds like you were seeing some problems. But yes, we do usually try to figure that out. It's really common that sites get copied on the web. And it's important for us that we can work out which of these versions should be shown in search and which ones we can kind of filter out. All right. So basically, the only step we can take is to submit the website to the spam report. And that's it. Yeah. I would still double check to see if DMCA works, just to kind of make sure that you're covering all of the bases. In particular, the report through Google, because that is something that, from our side, the legal team reviews. And if it's really one-to-one copy of your website, then that's something where maybe they'd be able to help resolve that. It's MCA? DMCA, yes. Oh, DMCA. OK, thank you very much. Sure. OK, is a category counted as duplicate content if it's indexed on Google? No, not necessarily. On a category page, you might have individual items that you have on your individual item pages. But that doesn't mean that it's duplicate content. I think duplicate content is kind of a tricky term anyway, because people often frame it as, oh, this is bad for your website. This will harm your website. Just because something is recognized as duplicate content or if parts of a page are recognized as duplicate content isn't necessarily a bad thing. It basically just tells us, well, we've seen this before. And if someone is searching for something in this block of checks that's duplicated, we need to find out which of the pages that have that copy of text are the most relevant. And it's not the case that a website will be thrown out of search just because it has some duplicate content. Because essentially, every website has some amount of duplicate content that's kind of, I don't know, it's almost impossible to avoid. We've been doing quite a bit of speed optimization on our page, and we managed to get a consistent over 90 score in our performance in Lighthouse. But this is only without our ad stack enabled. Once we turn it on, we drop in the performance scores to below 10. We've debated this at length with our ad provider, but they insist there's little they can do if we want monetization. We've done a bit of research and found that many face similar challenges with different ad stacks. So here is my question. Is anyone in the 90 plus range for speed in Lighthouse that has ads? It feels pointless to further optimize the page for speed with the limitation that the ad stack prevents. And then there are some examples. So I didn't check these pages explicitly. But essentially, when it comes to speed, we do take the pages into account the way that we get them. So if you have ads on these pages and the ads are the thing that makes the pages slow, then we see those pages as being slow. So that's something, I don't know, just kind of worth keeping in mind. With regards to the absolute score there, it's also worth keeping in mind that especially the 0 to 100 score that you have, I think in PageSpeed Insights, is something that's more of a guideline for finding different types of issues. And it's probably worth more looking at things like the Core Web Vitals, where you have a little bit explicit numbers to work on. But I mean, I don't know offhand which websites have really high scores and use ads. But the Chrome User Experience Report data, that is something that's public. And I believe you could even download that and you can double check different sites from your country and see how they are performing. You can also use the HTTP archive that also has the full Lighthouse Performance Reports in there. So you can look at things over time for other websites if you want. So that's something where, from my point of view, if you're worried about one specific ad provider, that's something that you can often test across different providers and see how other websites are performing. Just to add on that, because so far we haven't really found anyone in our, neither in our industry nor in our country region, that leverage ads and have reasonable Lighthouse speed, basically grades. So it seems really frustrating because we've spent so much time on optimizing. And really the only thing that seems to be preventing us is actually the only thing that makes us money on the page. So it's a pretty frustrating trade-off, to be honest. Yeah, I can understand that. That's always tricky. But I mean, on the other hand, the good part there is for the things that you're competing with, with other people, they have the same problems. So it's not exactly because we're also competing with government agencies that don't do ads. Since we're a provider, basically you have government weather services that don't need to monetize. OK, then maybe you have to be more creative. Now, I don't have any exact answer. So it's not from our side that we're saying ads are bad, and you should remove your ad so that you get a good score so that you can rank well on search. It's really just we need to make sure that when people search for something and they click on a result, they find the answers that they need on those pages reasonably quickly. That's kind of what we're looking at. And maybe there are other ways that you can monetize things in a way that doesn't prevent the kind of the primary loading of the page. So especially if you're looking at, what is it, largest meaningful content? I forgot what the names are. There's like so many. That might be something where maybe deferred loading of the ads would be an option. We are doing that. OK. It's actually been a large discussion with my ad provider because they always want to be in the head. And so we've been debating that. But still, we are getting these bad Lighthouse scores, which is surprising to me. But I'm also wondering if the scores are perfectly correct here. But really, we've been basically pushing back the ad stack as much as we can. Yeah. I mean, the score from 0 to 100 is kind of arbitrary because it's kind of a weighted score of different factors that are involved there. It's probably easier to look at exact scores which are based more on timing and those kind of things, especially when you're fine tuning the optimization there. Because things like, oh, was this served from ACDN or not, it doesn't really matter as long as it gets there very fast enough. I think you are also going to use those as a ranking signal in the future, so that has us additionally scared in a way. Yeah, I mean, we do use speed as a ranking factor on mobile already. It's more a matter of is this page reasonably fast or not. With regards to the Core Web Vitals, the new ones that we announced, they're part of the page experience score, which is kind of what we plan to use with regards to speed and ranking at some point in the future. We kind of haven't announced a date for this yet, and we'll announce the date at least, I think, six months before we actually turn it on, so that you have a bit of time to figure those details out if you feel that you need to do something there. And the page experience benchmark is based on, I think, a percentile from the site's point of view, so it's not that you need to optimize every millisecond, but rather it's like you need to be in the green range. And then if you're in the beginning or the end of the green range, that's more or less OK. Thank you. If I can intervene a bit, I ran the site through Lighthouse. So one thing that I think you can do, I ran the ads version, the one with the ads. So I don't see ads as being an issue in the sense that you're not showing above-the-fold ads at least on mobile, from what I noticed. So that shouldn't affect the largest contentful paint metric, which just that is like 25% weighted in the score of the overall performance score. But what I noticed is that you're showing a cookie kind of banner. So that shows up a lot later, and that affects how LCP, the large contentful paint, is calculated because your FCP, first contentful paint, is like one second, but your largest contentful paint is almost seven seconds, and that's only because it Google or, I mean, Facebook Insights Lighthouse calculates when that cookie pop-up shows up. So if you can. I think you're not seeing the rest of it because you're in a geo area where we don't have our primary ad stack. So you will be seeing something else than I am seeing because I'm in a region where there's a lot of stuff happening in the ad stack and you're just in the backfill mode. I guess you're not in Austria, I suppose. I tested with PageSpeed Insights as well. So I don't know if for the US you also have a difference. Now, so with regards to location, that's also a good point. We use, essentially, or the plan is to use the Chrome User Experience Report data, which is based on the normal users that are accessing your site. So one thing you don't have to worry about is whether or not people in the US have fast access because that's where Google is based, but really it's like where your users are based. That's kind of where you need to watch out for speed. And I guess it sounds like that's the place where it's the hardest on speed, which is kind of tricky, yeah. Andres, if you can start a thread on the Webmaster forums, I can happily take a deeper look into it and we'll define some suggestions. Sounds good. Thank you. Awesome. Cool. All right. Does having three open and closed HTML tags in a page negatively affect Google's ability to determine the main page content of a page? No. So this is essentially a matter of, is there valid HTML on the page or not? And that's not something that we would use with regards to ranking at all. The only time the valid HTML or not plays a role is if you have specific meta tags or specific attributes that need to be in the head of a page. And because the HTML is so invalid that the head of the page is kind of slipping into the body of the page, then we might not be able to confirm that these tags are actually in the head. So for example, if you have hreflang attributes on the page and the head of the page is so broken that we think it's a part of the body, then we would assume that these hreflang attributes are not actually valid because we can't find them in the head of the page. So that's kind of the one situation where valid HTML would play a role. But otherwise, it wouldn't play a role at all with regards to ranking Google. Let me just correct. One other place where that would play a role is when you're looking at things like structured data, where you need to have the HTML structured in a specific way so that we can extract the structured data properly. But that's kind of like if it's not, it would result in that structured data not being valid, which essentially would be the primary problem. It's not that the HTML has to be valid, but rather the structured data has to be valid. Is it possible for a domain to be punished if a site removes a lot of articles and makes them 404 due to some legal action or author requests? No. So removing content is essentially, it is perfectly fine. It's part of the normal web. The only kind of punishment, if you will, that you will see is that obviously, once this content is removed from our index, you won't rank for that content or for the content that used to be there. But it's not that the rest of your site's content would be ranked worse just because you have a lot of 404 errors. 404 errors are essentially part of the normal web. It's not something which we would take as a sign that the website is low quality or bad or anything like that. Could the change of address tool be used in some way for migration where you're consolidating two domains down to one? Trying to think outside of the box on this. I could see a situation where possibly you could use the change of address tool like this in the sense that you could probably set up the configuration of your sites to make it so that we would process the change of address tool there, because there's some requirements that we look out for with the change of address tool, such as that you have a one-to-one redirect from the old site to the new site. But you could theoretically set something like that up. Whether or not that makes sense, I don't know. Because the tool is really meant to transfer essentially all of the signals from one site to the other site. And the idea is to replace the new site with the old site. And if you're consolidating multiple domains, then it's not that you're replacing the new site with the old site's information, but rather that you want everything added together. And that's something where I could imagine you might be able to use a tool from a practical point of view. You can access it. You can click OK and make it kind of activated. But I don't know if you would have any positive effect over just redirecting normally. My feeling is probably you would see some weird effects. But I don't know if, overall, you would have something positive out of that. We fixed some really old product URLs with 301 redirect and see some huge spikes in backlinks. How does Google react to this? We're seeing some substantial decline in ranking after that. Is it related? I usually changing URLs on your website, fixing old URLs, redirecting them to make them valid, that's perfectly fine. That's not something that I would say would be problematic. It wouldn't be something that we would say is a sign of spam, definitely not, because you're improving your website. My feeling is the decline in ranking is probably unrelated to fixing the URLs on your website, but maybe something that was happening anyway. And it's kind of independent of that. So I would continue to improve the URLs on your website, fix them with redirects if you notice that they were kind of linked incorrectly, those kind of things. If there's a domain whose archives shows some different themes and content than the current theme and content, should we consider that kind of domain as good for backlink purpose? So I don't know what you would consider good for backlink purpose. That's kind of the part where I'm a little bit stuck. But essentially, just because the cache page looks slightly different than the current page, that's not a bad sign. That's usually a sign that the website is being worked on, that things are changing. That's perfectly normal. With regards to good for backlink purposes, it sounds like you're dropping links on other sites. And that's kind of the bad thing. It's not so much that the website is being worked on, but rather, if you just go out and you add links to other people's sites, that would be kind of not so cool. Since a couple of months, the fabric on of my site disappeared from mobile search results while it keeps being shown in Search Console. I checked the Google guidelines, and everything seems OK. The only thing that changed in the same period is that I changed the URL structure from www.site.info to www.site.info slash em, since I have multilingual versions, all with their own site map. So the root is no longer being indexed at 301 redirects to slash em. Could that be the reason? I don't know. That's actually an interesting question. I haven't looked into how that would be handled. So I think on the one side, one of the things I've noted with Fabicons is that it just takes a lot of time to be processed. So if you make changes with the Fabicon, if you fix something, if you change the URL, then that's something that takes quite a bit of time to actually be reprocessed. So if you made this change, I don't know, a month or two ago, it's possible that that's still being processed. On the other hand, it feels kind of weird that if you don't have a root URL, but rather kind of like you're redirecting to some lower level page, if that's the only change that you made, then that seems like something that we should be able to deal with. I am online. OK, cool. Can you drop me a link to your site, maybe in the chat? And I can pick it up afterwards and double check with the team. Yeah, yeah. Because one of the requirements for the Fabicon is that the root must be crawlable by Google. And my actual root, so that the domain is not indexed because I have two home pages, one for one language and the other for the other language and so on. So the basic, if you type the basic home page, you get a 301 redirect to the default language that is the English version. And so actually the real root is not crawlable by Google, but the English version. Yes, so that was the question, if it is this an impact. Yeah, yes. I don't know. It's the first time I hear about this configuration, but I'm happy to take a look with the team. Is it a bad configuration having just home page on different language and not make the domain index? It's essentially fine. It's not something that I would say is problematic. I think sometimes it might confuse tools if you're running tools over the website and you don't really have that root home page. But from a search point of view, this is something that a lot of websites do. So one thing you might want to do is look at hreflang and use the X default for the primary version that you have. But that's, I don't know, kind of like small optimization. It's not something that is critical or that you need to do. I will drop you the line. Thanks. Sure, thanks. I want to know that I update my website every day with a new post, and Google indexes the post, but not the keywords of that post. So can I know the reason behind that? I don't know. It's hard to say just based on that question. So if we're indexing the content of a page, then we essentially have that page indexed, and we could show it in the search results for any searches that are made for that content. What might happen, or I guess, depending on kind of the keywords that you're targeting, is that if you create something completely new, that it doesn't immediately rank for kind of competitive keywords. So that's something that feels kind of normal from that point of view. So maybe it's a matter of just taking more time, working out ways to improve your content overall, and kind of growing your website from there, rather than just creating new content every day and hoping to rank for the keywords that you have in that new piece of content. But it's not something where I would say we have any systems on our side that say, oh, we will index all of this content, but we won't show it for any of the keywords for it. Because if we want to kind of index it, then we want to index it with the ultimate goal of showing that in search. It doesn't make any sense for us to index something that we wouldn't want to show. My website is in Marathi, Indian language, and will it get indexed for the English translated queries? So if your website is completely in one language, then generally, we wouldn't be showing it for queries made in other languages. There are always, I guess, some exceptions to this and some situations where we do kind of this kind of thing where we recognize that someone is searching in one language, but we can't find any content in that language, and therefore, we will translate that query into maybe another local language and try to see if we can find something there. I believe that's something that we've done. I don't know if we still do it in India in particular, where maybe there is not a lot of local content yet, but if we translate the query into English and we can show English pages using Google Translate, perhaps, then that's something that might make sense there. But in general, if you're working on your website and you want to rank for queries in a particular language, then I would strongly recommend that you actually also write your content in that language. Using things like hreflang lets you connect those different language versions, but essentially having content in a particular language version, if you want to be shown for queries in that language, that's really important. And that's especially important when it comes to more competitive terms, where there is a lot of content already in one language, then you really need to make sure that you're putting your best foot forward, that you're creating the content as ideal as possible so that, on the one hand, our systems can look at it and say, oh, this matches exactly what the user was looking for. And users, when they go to your pages, they're like, oh, this is fantastic content. This is something I need to recommend to my friends, for example. My website is only restricted to the US demographic. Can I bring links for my site from outside of the US country? Again, can I bring links to my website? That sounds a lot like I would like to randomly create links on the web pointing at my website. And I want Google to think that these links are natural, which is more the problematic aspect there. With regards to restricting a website to the US, that's something you can do. In general, we crawl websites from the US in most situations. So if we can access the web page, we would be able to index it. I think it's a bit, I don't know, tricky or not always a great of a user experience when people outside of the US click on a search result like this and they essentially can't access the website. But sometimes you need to do things like this for legal reasons. I guess the problematic alternative to this, just to have mentioned that as well, is if you have a website that is being crawled from Googlebot in the US and you want to block users in the US from accessing the page, then Googlebot wouldn't be able to access it either. And we wouldn't be able to index it. So that's kind of if you were in the other situation where you needed to block users in the US. But with regards to links in general, if people are linking to your website, then that's something that we take into account, regardless of where those links are located. From for multilingual block, which is best subdomain or subdirectory and how to configure that in Search Console? Oh my god, I don't know if I want to start this fight again. Essentially, when it comes to multilingual websites, we don't care what the structure of your website is. So that's something where if you feel it makes more sense to do it on different subdomains, then go for it. If you feel it makes more sense on subdirectories, go for that. If you want to use URL parameters, that's fine too. So for different languages, any URL structure works as long as you have one language per URL. So as long as you don't have this fancy setup that your website on the same URL returns with different language content, depending on the user location, because we would not be able to index that. On the other hand, if you are targeting multiple countries, so not multiple languages but multiple countries, then we need to have a clear structure of the website for individual countries, which means, again, subdomain or subdirectories would be the approach to take. And in Search Console, you would need to specify or verify those separate sections and specify the geo-targeting of those sections. So for multiple languages, any structure you want is fine. For multiple countries, you need to use subdomains or subdirectories or different domains, even, if that works for you. I have a client that has one of their navigation options as a JavaScript dynamic URL. So if someone clicks on media in their navigation, they land on slash media page. But all the resources and the blogs live on slash media type equals blog, et cetera. I know Googlebot is smart when crawling JavaScript, but some SEOs say it's best to have clean URLs with past like slash media slash news. How good or bad is it to have the navigation this dynamic for SEO? I also read about hash bang on Google's blog. I feel my lack of computer science made this topic a blur to me. So on the last part, that's perfectly fine. We strongly don't recommend using hash bang anymore. So that setup is something I would not bother looking into. You don't have to worry about your computer science skills for that. With regards to dynamic URLs or not, I think, as I understand it, the problem is less that it's based on JavaScript, but rather that you have the question mark in the URL and then you have type equals blog or different parameters within the URL rather than just files and directories. And from our point of view, that's perfectly fine. That's not something that you need to change. So these URLs with parameters or URLs without parameters, from our point of view, they're essentially all equivalent. They rank exactly the same in search. Sometimes it's even like if we talk with the crawling and indexing teams at Google, they tell us people should not try to hide these kind of parameters because having parameters in your URLs makes it a lot easier for our systems to actually crawl and index them. And the reason for that is that we can recognize that these parameters are unique elements of a URL. And our systems can learn, does this parameter make sense? Is this something that is critical for crawling and indexing of the page or can we remove it? And by learning these kind of things across a website, it's easier for us to crawl more efficiently. So if you have kind of everything in a path of a URL, then it's hard for us to say, well, this path element, the fifth one in the line, is something that we can drop or kind of simplify for crawling and indexing. Because that gets a part of the path. And theoretically, it's something completely different. Might lead to a different script on the server, all of that. However, if it's a part of a parameter in a URL, it's like question mark type equals media or I don't know, page equals 2, then that's something that our systems can learn and they can understand that this is critical for the page or this is less critical for a page. And we can optimize our crawling and indexing based on that. So it's definitely not necessary to go from parameter URLs to kind of like just pure file name or directory URLs. I think sometimes it makes sense to have cleaner URLs just for users because sometimes users copy and paste URLs and they share them or they link to them. And having really nice and clean URLs makes it a little bit easier. For the most part, I feel URLs themselves in search are a little bit overrated because they're less and less visible in the search results. In browsers, they're less and less visible as well. So that's something where things like using breadcrumb markup will probably have a stronger effect with regards to the visual identity in search when it comes to individual URLs. So I guess I drifted off a little bit on a tangent there, but essentially having these kind of parameter URLs or having clean URLs is essentially fine. There's no need to kind of force it one way or the other. OK, we're kind of running towards the end and tons of questions left. Like I said, I'll try to go through these and add some comments as well on YouTube for all of the things that we don't get to. But if any of you want to kind of jump on in with more questions live, feel free to do so. I also have a little bit more time afterwards if any of you want to stick around and maybe ask some questions off the record, I guess. Hi, John. Are you able to hear me? Hi. Yes. Yeah. Actually, I wanted to ask more about 301 and 404. We have got more than 1,000 pages in our website, that is 404. And we would like to solve them. And those pages are also getting linked back from external websites. And I wanted to know how we can solve that thing because that is out of our control because they are external links. And what should we do? Should we add 301 directs to all 1,000 pages to the relevant pages? Or whether, I mean, I also, my concern is whether this will increase our crawl buzzard or not. So if they're already returning 404, then we essentially don't use any of those links to those pages. It's not that they cause any problems. Essentially, you can keep it like that. If you've replaced what used to be there with new pages or with different pages, then using a redirect is the right approach there. So for example, if the old page is one product version and then you stop selling that product and it went 404, and now you have a new version that replaces the old version, then redirecting from the old URL to a new one is perfectly fine. I think that's good for users. That's good for search as well. What will happen then is any external links that we're pointing at the old version, they will be forwarded to the new version. And that's essentially OK. But just going in and saying I have 1,000 404s, I will redirect them all to my home page. That's something where our systems, when they look at that, they will say, well, it doesn't look like your home page has replaced these pages. It's more that you're doing this as a way of having a 404 page that just shows users a home page. And in a case like that, we would probably treat that as a soft 404, which means internally we would say this is essentially the same as a normal 404, and we would just drop those pages and drop those links or anything that was associated with that. So if you have a replacement, user redirect, if you don't have a replacement using a 404, is perfectly fine. And what about the call budget, whether it will be the call budget of our website? I don't think it would change anything. Because if it's a 404, we would probably crawl a little bit less often over time. But if it's a 301 that we treat as a soft 404, for example, that's all the same. I don't think you would notice any big difference. Even if the website had millions and millions of pages that were like this, I don't think you would notice a big impact on crawl budget. OK, thanks a lot, John. Sure. I can ask a question? Sure. Is there any negative impact on blocking with the robots.txt resources that we think they are not to be indexed by Google, like the cookies, JavaScript, some other pop-ups, or other stuff that we think it's not worthy? So we block with robots, or Google is not so happy that we cannot download some resources? Ultimately, that's fine. What's important for us is that we can render the page and see, for example, that it's mobile-friendly or not. So if, by default, you're blocking all of the CSS and all of the JavaScript, and that makes your HTML page kind of unreadable on mobile devices, then we would think this is not mobile-friendly. But if we can render the page and we can tell, oh, this is a normal mobile page, and it's just like one JavaScript file or one data file that you access, you need to block it, that's perfectly fine. Thanks. Sure. John, can I ask a question, please? Sure. It's a question about site links. It was actually posted in the comments. But is there anything we can do as webmasters about the text that is underneath individual site links? Because it's throwing out gibberish for one of my clients and they're querying it, so. Oh, no. OK. If it's really gibberish, it would be great to have some examples of that. Usually, site links are based on the structure of the website as we can recognize it. So that means it's something where we maybe found a link that was going to one of those pages using an anchor text like that within the website. That's something where if you're seeing pure gibberish, then it sounds like we're not able to recognize maybe the anchor text on some of those pages, which could be maybe something weird with regards to JavaScript. It could be maybe something weird with regards to the encoding of the characters on the page, especially if it's something in, I don't know, non-English or non-the seven-bit ASCII character set. Then sometimes it can happen that we pick up the encoding wrong. And then the anchor text, obviously, will also be wrong. What I've also sometimes seen is, depending on how you link pagination within a website, sometimes we'll just take number seven and use that as a site link because we think this is particularly relevant. Because maybe you're linking to all paginated pages from all pages of the website, then we think, well, number seven is as good as number five. So maybe we'll just link with number seven. That can also happen. But if the internal linking is really fine, if you use kind of a crawler of your own to crawl the site, you can tell the anchor texts are all OK, and we're picking up something that really doesn't make sense, then that would be something to pass on to the team. OK. I'll double-check all that and send you a link or a screenshot or something. Cool. Sounds good. Thanks. Sure. All right. Getting kind of quiet. OK, so maybe I'll just pause the recording here. If any of you want to, you're welcome to stick around for a bit longer. Like I mentioned, I'll try to go through some of the other submitted questions and add some comments where I can to catch up a little bit there since there's so much stuff. But in any case, I hope you found this useful and would love to see some of you again in some of their future Hangouts again. Thanks for joining, and wish you all a great weekend. Bye.