 OK. All right. Welcome, everyone, to today's Webmaster Central Office Hours Hangouts. My name is John Mueller. I am a Webmaster Trends Analyst in Switzerland. And I'm here with a bunch of awesome other people who are joining us live in person for the Hangout. Do you want to introduce yourself? So my name is Eduardo Borges. I teach SEO in Portuguese languages, so Brazil, Portugal, Angola, Mozambique, and some other countries. And a pleasure to be here in live. My name is Christian, and I'm blogging mostly in German on the German blog, SEO blog. And yeah, great to be here. Thank you, John, for inviting us. I'm Jenny. I'm an in-house SEO at the marketplace for moving, cleaning, home services. Yeah. And I'm Aurora. I manage the English and the Spanish forums for Webmasters. Ooh, wow. So many cool people here. And we have a handful of people who are watching here live as well. Good to have you all here, too. Sometimes it's tricky with introductions, but if you want to, feel free to introduce yourself briefly. I'll introduce myself, since my camera's not working anyway. I'm Rob. I own Experience Days, who's had the worst Google issue in John's career. There you go. And he still owes me an explanation, and he's trying to avoid me, but he can't. Next. Hey, guys. I mean, hi. I have an SEO agency here in Romania. I also participate on the Webmaster forums. And I've also created the Search Analytics for Sheets add-on for Google Sheets, if anyone has used it before. Yeah, that's about it. Hi, I'm William from Taiwan. I'm working as an SEO in a travel platform. I have to manage seven languages, and lots of questions about John, lots of language issues, which I give them, indexing, and problems that I have to solve. Next. Hi, I am Saidu from Sydney. I work for the only marketing agency named Real Clicks. Hi, I'm Gev from Armenia. I'm representing a startup called Ucraft, where a website builder startup. And I'm doing SEO for my own startup with my team here. So I'm happy to be in this call today with you. Cool. Anyone else? I think that's. Can you hear me? Yes. Yes. I'm Sanjit from India, Bangalore. Actually, I work for a health care portal, Prakto. So I manage Azure for a Prakto. Cool. All right. So as always, maybe it makes sense to start off with a question from the room. What all is on your mind? What should we talk about today? Yes, nothing, sir. I do have questions about RankBrain. RankBrain? Yeah. OK. So the question is, I want to try to identify the difference between SEO in Portuguese and SEO in English in terms of RankBrain. So for example, we have a word like cookie in Brazil. You can say cookie like three different words for that, bolacha, biscoito, bolachinha, biscoitinho. So that'd be variation here. So I'm trying to understand if the algorithm for RankBrain works the same way in English. Sorry, John. The microphone is mute. We can't hear you. I think someone needed you. Yeah, someone just needed you. You? OK. Someone might have accidentally picked your wrong button. OK. Let me start over. My question, again, is regarding RankBrain. I want to know the difference between Portuguese and English, languages. So for example, in Brazil, if you want to say cookie, you can use four different words to say exactly the same thing. And when a client or a student wants to optimize their page for that, he asks, OK, should I put the bolacha and biscoito, which is both words in the URL, title, and all that stuff? Or should I just trust Google, RankBrain, that he's going to identify that and just show the user what is right? We know that RankBrain knows how to do that in theory. But I want to see the difference between Portuguese and English. The database, is it as big in Portuguese as in English? Like the person page of the searches? I don't know. So I think in general, we do try to recognize when things are synonyms, which we try to do automatically. So it's not that someone is writing all of the words that they know in these languages and saying, this matches this one and this matches that one. It's more a matter of our systems learning this over time. And that's something that sometimes you notice when a new word comes into the vocabulary, where it becomes really popular, and everyone uses that word for this specific topic, then you see in the beginning, it's treated as something completely separate. And then over time, it learns that actually, this belongs to this other word or this other topic. So that's something that is learned over time. It's not that we have like a specific database where we say, this is how many words we have and how we treat them. And between the languages? Between the languages, it's trickier because we tend not to have a direct link between the languages, where we see someone is searching in one language and we think we should show them content from another language. We tend not to do that. I think it might be tricky with places like Latin America where you have, like maybe you have Spanish, but it's not the same Spanish as you have in Spain. And in those cases, we will probably try to learn those as synonyms because we see they have the browser set to Spanish and they're using them interchangeably. So probably we will learn that. I think with regards to what to put on the pages, that's something you almost need to think about your target audience. How do they search? What do they want to find on the pages? What makes them comfortable when they go to that page? Is it like, this particular writing of the way or do they search for a different type of writing? Because ultimately from the SEO side, we do try to figure out which ones belong together. And sometimes that works, but if we see that a user is searching explicitly for one version and we know your page has that version, then that's already a good thing. And if they go to your page and they see, oh, this is what I was looking for, then it's also a good sign that indirectly we could pick up on over time. So would you say it's a good practice to use both synonyms in the page? If you see that people that you're targeting are using both of them, then that might be an option. You know why I'm asking? OK, so some ball, we say bolasha. Rio de Janeiro, we say biscoito. And both are trying to buy cookies. I think that can be fine. No, putting both in the page. I think that can make sense. Yeah, I think you have to be careful that you don't go off into the typical SEO keyword stuffing thing where it's like, oh, I will put plural and singular and all of the different variations of the words on a page because that doesn't really help. But just being direct and saying this is what people are searching for, this is what I offer, I think that's OK. Sure, thank you. Yeah, anything else on your mind? I've got two other questions concerning the upcoming speed update. Speed update. Yes, in July. And I would like to know when exactly it will begin to roll out. And the second question is, what is a page speed or loading time that is enough for not getting worse rankings? I don't think we announced an exact thing. So that's something you probably have to watch out for on Twitter. Like, suddenly everyone will go crazy online. But I think with speed, it's something that you don't just turn on in one day. You have to work on it over time. And then when it's ready, your site is faster. We don't have a specific metric that you can focus on where we say, this is the number you need to get to this particular area, mostly because we think that you shouldn't just focus on one number. You should look at the overall picture and maybe look at the different tools that have different focus areas for speed. And it's a bit different than the other speed update that we did in general, maybe seven, eight years ago, where there is no differentiating between fast and slow. It's more of a gradual scale. So the faster you can make your pages, the more we can take that into account. And it's not so much that it's too slow or it's fast enough. OK, so gradual improvements can help me with my rankings. A little bit. OK, cool. Anything in your mind so far? It's a different topic. I mean, there's been a lot of talk about search, like voice search, and people searching for queries like near me. And I've noticed, or I found a website in the US. And they added near me into every URL and also on a page. And I mean, it seems to work for them. But how feasible is that? And if suddenly everyone adds near me and the URLs, I mean, that's a bit weird, I guess. I mean, it's tricky because on the one hand, when we see someone searching for near me, we try to turn that into we know where their location is and find local results nearby. And not so much that we look at, oh, near me has to be on the page. Because if your website is in Zurich, then having near me in the title doesn't make it relevant in Germany. And that's something where I suspect there might be some things where you can get some value out of having near me in the titles. But it makes more sense to really be explicit about your locations on your pages directly, so to have the addresses on there, to have the Google My Business listing, so that we really know when someone searches near me, that means near this particular location in Zurich or wherever. And then we can find the websites and businesses nearby. So I think it's not that sustainable to put near me into titles everywhere. But if it works for people, nobody's going to complain. I don't think anyone from the web spam team would say, oh, near me is that title. If you want to have it, fine. Hi, John. Hi. We have faced a problem with one of our websites recently. The problem is we got a message from Google Webmaster tool that we have a mail order on our website. We got the message on 14th of June. So when I got the message, they gave me a sample URL, which has the issue. And that URL, actually, is one of our landing page, which we created for our PPC marketing. And that is actually one page landing page, and it has a link to our main website. So I removed the link and fixed it. Then 24th of June, I got another message that I had mail order on my website. And I checked my website. I did not find any issue on the website. And when I click on the download sample URL, so when I got the message, there is an option for download sample URL to see which URL has the issue. When I click on the download sample URL, I download the CSV file. But the file was empty. There was no sample URL. So I was confused. What is the issue? Because the message is telling me that I have mail order issue, but the file has nothing. So I click on the review. I request to review the website again. And I got a message. It started that the problem is fixed. They review my website successfully. But today, I got the same message again that I have issue mail order on my website. I click on the sample URL option. I download the file and file is empty. So what is the issue getting this message? I suspect the malware type thing that is flagged there is the uncommon downloads. Is that possible? Does it say something like that? No. We don't find anything. I'm going to search everything on the website. We didn't find any script, anything like this. There is no redirection, nothing. OK. So one really, well, I don't know, really common. So one thing I've seen recently is the uncommon download warning that we have with regards to, I don't know, problematic URLs that we found on the website. And we alert webmasters about that in Search Console 2. Uncommon downloads basically just means we haven't been able to check the files that were downloaded. And we show a warning to users when it's completely new. So if it's a new download that you have on your website, a zip file, or some executable or something that is really new, then that might be flagged like that. And in those cases, we don't really have sample URLs because we've seen people just kind of abuse that system and say, oh, I will delete that file and put a new one up there instead. And that causes more problems. But we don't have any file on our website like this. No downloadable file on the website. The website is kind of normal text-based website. No. So what also might be the case is that there's something problematic with the security of the website in general that someone else has placed files there, which might be a bit harder to spot. Because sometimes they're very tricky and that they cloak things from the webmaster and they hide the log files. And we'll send you a warning about things like this. But as a webmaster, if you're not used to digging into security issues, then it's really, really hard to find. So if you're seeing things like this where you can't find the source of the problem and it's being flagged as malware, I would recommend going either to an expert directly or going maybe to the webmaster help forum where there are lots of people that have a bit of practice in digging into what exactly might be happening. OK, I have another question. So this is from one of our clients. Like he has an e-commerce website and they remove product from the website, adding product to the website. So we need to get a lot of Q1 redirection. When they remove the product, we add a redirection. Sometimes they change the URL. Sometimes they change the URL structure for most of the products. So we need to add Q1 redirection for the website. So is it bad if a website has a lot of Q1 redirection? If it has a lot of keyword variations. I think you have to be reasonable in the sense that you should look at what people are actually trying to find on your website and have that kind of content on there. And just adding variations of different keywords usually doesn't provide a lot of value. The downside of adding more and more variations is also that you dilute the value of your content. So if you take five pages and you turn it into 50 pages, like all of just variations, then those 50 pages, they have to fight on their own. Whereas if you have five pages that are really strong, that's more likely that we can show them in search. So that's something that's worth balancing. Sometimes it makes sense to split things a little bit. But a lot of times it makes sense to just have really strong pages. OK, thank you. All right, so let me see what kind of questions were submitted. And we can have more questions and more discussions. OK, awesome. All right, so our website ranks for very few competitive keywords, including free logo maker. We have naturally grown strong backlink profile for this page. The question is, we want to start ranking for logo maker, which is slightly difficult for us because Google thinks our logo maker is only free. Do you think we should change the URL to just logo maker instead of free logo maker? In that case, if redirected properly, will the experience drop? I think if your logo maker is not free anymore, then taking the word free out of the logo maker URL probably makes sense. It's a little bit less confusing. But in general, any time you go from something quite specific to something more general, then it gets harder with regards to search. And I think this is a normal kind of path for a website to start with something small and be really good there and then expand from there. So I think this is something probably worth trying out and seeing how far you can go. It's not always easy, especially if you go from one very specific topic to a very broad topic or a topic with a lot more competition. But I think it's a good approach to take to expand from one place to another. And I wouldn't worry so much about Google thinking that your website is always free. Therefore, it will never rank for something commercial. Google presents phone numbers for companies at the top of organic listings. What should we do when these go disastrously wrong and start showing contact information for the wrong company? I spotted your tweet when you sent that. And I sent it to the team here to take a look. In general, I think there's a feedback link on all of these search elements. So you can also let us know that way. If this is your company, then you can give us information about that, I think, through structured data where you can tell us which phone number you want to have shown. But if it's not your company, then obviously you can't change their website. So that's a big figure. Thank you for doing that, John. Appreciate it. My chat site, the word chat, an awful lot. Can a keyword stuffing penalty cause a site to be demoted for a particular term? My pages rank top five for tons of desired terms. But I'm nowhere to be found for the one I should be ranking for, which is free chat. So it's certainly possible that our algorithms see a lot of keywords on pages like that and that we think, well, maybe this is way over the top and that our keyword stuffing algorithms will essentially say, well, we're just going to ignore this completely. So that could certainly be happening. I don't know what your page looks like and how often you have that on there. But it's something I see very rarely, but it does happen. Where you go to a page and you look at the source code so you can see all of the alt text and you search for their words. And it's like 300 times on the page, OK. This is a bit extreme. So that's something where maybe it's worth taking a step back and looking at how often you actually do have that on the page and if you really need it that often for normal users. Because search engines are generally pretty good in that if you have something on a page maybe once or twice, then we'll pick that up and know that that's relevant even without having to repeat it hundreds of times. I think there was a question from one of you. Hello, John. Hello, John. Hi. John, I had a question about the site being responsive versus on the m.domain. So if there is a scenario in a case where, say, part of the website is on the m.version and the blog or the second part, as I call it, because it's on WordPress, is on a responsive version, how good or how reasonable is that as an architecture? And does it have any implications in today's world? I think that's perfectly fine. So we look at mobile versions on a per page basis. It's not that we say the whole site is m. or the whole site is responsive. It's like we see this page has a responsive version. This one has an m.version. That's perfectly fine. I think with mobile-first indexing, when you have separate mobile URLs, everything gets a little bit more complicated. Because you have two pairs of URLs that you have to watch out for. In the past, we would only focus on the desktop version. But with mobile-first indexing, we're focusing on the mobile version. So if you use any SEO tools to check your sites, then you need to test both the desktop version and the mobile version. Whereas with a responsive site, if it's completely responsive, then you just check one version. It makes it a little bit easier. Sure. Just a follow-up question on that. So like in the normal implementation, we add a canonical tag of the desktop on the mobile version, saying that this mobile m.version is equal to the desktop version. In this mobile-first world, should we do the opposite? I'm not sure, but just a doubt. No. You don't have to change anything with the canonical tag. You can leave the listing link with the link alternate to the mobile version and link canonical back to the desktop version. And that works for us for mobile-first indexing, too. We explicitly worked with the team to avoid making it so that webmasters have to make changes there, because that's always a place where things go wrong as soon as you have to think about when do I have to change things and when my site gets switched to mobile-first indexing and you don't know when that will happen, you can't expect people to change that. Sure. Thank you so much, John. Sure. John, I've got one question. All right. Go ahead. Sorry for interrupting. I was the person that asked the question about free logo maker keywords. You read that question at the beginning, but then it got disconnected. I wasn't able to hear your answer, but I assume that you answered it the way I think. But I've got another question. The question is about the following. Let's say with free logo maker from United States, we're ranking in the top one or two, first or second position. And with just logo maker, we rank on 20 position, 2-0. So the difference is really, really huge. So what I'm thinking is just remove the, I mean, change the URL, as I mentioned in my description in the text. And the question basically is, during this time that Google will realize that, hey, these guys have changed the URL, correct, redirect, will he still be sending the huge traffic that he is sending with that keyword? And then will we see a drop in the ranking for free logo maker? I don't know. I assume not so much. But if you take away everything that refers to the free aspect, then over time, our systems will think maybe it's not the perfect result for free logo maker anymore. Maybe it's more relevant for just logo maker. So that's something, if you change the focus of your website, that does take a bit of time to be reprocessed. And you might see results if you just expand from your existing to add a page about just logo maker or, I don't know, cheap logo maker or something like that. Then that makes it a little bit easier for us to say, we'll keep everything around free logo maker the way that you have it, because you still have it on your site. And if you expand from there, we have new content that might make sense for this broader term. Yeah, but the most frightening thing here is that most of the websites all over the world are linking to our website with an anchor text free logo maker. So this is very frightening for me to just switch to just logo maker, because Google will say, hey, everyone thinks that this is free. I don't know. It's really frightening. Yeah, I think it's a strategic decision that you have to make. Do you want to change the focus, or do you just want to broaden? You want to add additional parts. And that's something I don't think we can solve for you. OK. OK, thank you. Sure. So again, regarding Portuguese and English. So when a website in English points a link to my blog in Brazil, in Portuguese, does the page rank or the power of that link would be the same of another blog, like Saint Quality, pointing the same link, but in Portuguese? What I'm trying to do to find the anchor, does Google translate to know what is about, or just find the niche and just send it? We don't translate the anchor text to see what that would be in another language. But if it's a link, then we can pick that up. The anchor text, usually when we see different anchor text for an individual page, it just helps us to understand the broader area a little bit better. So it's not that we would try to translate it and say, oh, this match is this word exactly. It's more that we understand, well, it's not just this word that's on the page. It's also this bigger topic that people are linking to this page about. OK. So for example, a travel blog in Portuguese. And now we try to promote. So we promote. We can choose promoting only to Brazil. And now with the DMCA rules, we're like, OK, so do you think it's worth it promoting in other countries as well? Because maybe the link's not that strong. So should we just focus 100% in Portuguese? Would you say both are good to be promoted at, or should I focus on my language? I think it's fine to promote it. What do you think people are that would like to go to your website? But for SEO reasons, you think the balance? I wouldn't worry too much about that. OK. No. I think that should just work. Cool. Anything else from your side? And there was research recently telling that company blogging doesn't have so much value for SEO aspects. I don't know if you read it. It was published by Cystix, quite popular in Germany. And my question would be, if you are blogging, you have this timeline. And your articles go down sometimes on pages that are lying behind. So they are not directly accessible from the homepage. And so the internal linking is weakening over time for these blog articles. Is this a problem concerning rankings in Google? If I have a very old blog article that's maybe on page 25 in my blog timeline, or is it OK? I think that's OK. I don't see it. So I didn't see this article. It sounds like it was on the Cystix company blog. So that would be kind of, I guess, almost ironic if they say. Let's go. I don't know. But in general, I think having a presence online that provides a little bit more context around what you do is always a good idea. And it's not necessarily that you need to promote your core business all the time like that. But it gives people a little bit a bigger understanding of what the company is like, what they're focusing on, what they're thinking about, maybe some background information. And with all of that gives some things that make it possible for people to find, like the blog or some general information about the company through search as well. So it's maybe not driving so much traffic for their core keywords that they're focusing on, but more for just everything kind of around that business, which I think can have a little bit of value as well. You don't have to do everything just for SEO reasons. That's why I think if you have a good company blog and it's interesting for people and it helps your customers to build a better relationship with your company, then that's fantastic. Even if there's no SEO benefit at all. It's a great place to place your updates, for example. We do that all the time. And users appreciate it. Yeah, you do it for the users. And in this way, not for SEO in the first place. Exactly. OK. I think if you can do something to kind of build that bond with your users so that they appreciate your company, they search explicitly for your company, that you don't have that much of a problem that you need to rank for the generic keywords because people are saying, well, I really want to go to this company for this topic, then they'll find you automatically in search. Right? Could you just on that topic, I see a lot in any of the forums I go on and people I try and help, that they seem to separate blog from website as if they're two different things all together. They're not even remotely related. But can you just address the kind of facts in terms of, a blog is just a website with pages. It's not. There's no, apart from maybe the CMS that might be different because you're using an off-the-shelf system. If you could address what the difference is of any. But I keep telling people, it's just a web page. Just do what you would do. Yeah, that's perfectly right. So that's something I do here every now and then. Like, should I do a blog or should I do a website? And for us, it's just a URL and it has content on it. And the technology behind it is totally up to you. A lot of the blog setups are done in a way that they just work for search. They work really well. They have feeds. A lot of times they have standard themes that just make it possible for us to crawl them very easily, to find the content on them. So it's a very simple way to put content out there. So I definitely wouldn't shy away from using a blog instead of a traditional HTML website. Because in the end, we look at those pages and they have the same content and the same thing for us. We don't have a classifier in search that says, oh, this is a blog. Therefore, we should not treat it as important as, I don't know, handcrafted HTML page. I also don't think users really differentiate if they're searching for the same thing. They want to see the content. They don't care what the technology is behind it. All right. Thanks. Hey, John. So since we were talking about some of the context around the links, so I won't go into that one. So we have this PR agency and we do a lot of announcements around our product, brands, and everything. So at times, they reach out to the media outlets and they publish it also. So at one time itself, we might be having a lot of news stories around our brands and everything. And sometimes they link to us also. So is there a kind of a to worry? I mean, if we're having too much of links coming from those different publication websites. So is it kind of problematic or is it like, OK? So we try to ignore links from things like press releases because we know in general companies put the press releases out themselves. So any links in there are essentially placed by themselves. But if these links happen and they're out there, it's not something you need to worry about because you can't really take them all back. I just wouldn't rely on kind of press releases as a strategy for building up links for a website. Because like I said, we do ignore most of those. Thank you. What about niches when people are using techniques to grow their backlink profile, like plaque head techniques? Is there a way I can just, I don't know, do something so I can, like, my good content goes rank against those? It's, I don't know. So I guess there are different aspects there. One thing you could do is if you see people doing something really problematic, you can tell us in the web spam report form so that the web spam team can take a look. That's something that I think makes sense and is really useful for us, too. If we get these reports and we can take action on them, or at least if we can learn that there's this bigger pattern of things that we're doing incorrectly or not catching correctly, that's also useful for us to kind of look at the bigger patterns. But otherwise, I would mostly just focus on what you would normally do. So it's always, I think, kind of tricky. And it's worrisome when you see your competitor doing something bad and they're ranking above you. And you're like, ah, Google is falling for their stupid tricks. But instead of focusing so much on what they're doing, I would really just kind of spend that time on your site. Because everything that you do on your website is sustainable. It's something that lasts over time. Whereas if you spend hours every day reporting your competitor doing this bad thing and this bad thing and this bad thing, then that's all time you lost that you could be putting into your site. So the web spam report works? It works. It works. So it's not the case that the web spam team takes these reports one by one and always just blocks that site completely. We do take a look at a lot of those reports. And a lot of times, we see that the algorithms already ignore those tactics. And sometimes we see that something is still slipping through. And sometimes we see that there's a bigger pattern of things that we need to take action on. So I wouldn't assume that you can submit something there and then next week that website will disappear. There's a lot of trying to figure out which of these reports are useful and relevant before we take action on them. And sometimes the action that we take is to ignore that particular tactic on the website, which doesn't mean that the website will disappear from search. Is there like a period? Like you say, you can take six months. You can take a week and one year. It really depends. It really depends. And I think it's, I understand it can be frustrating. Like if you're a competitor, you look at them and you see they're doing something sneaky and they're ranking above you and you're like, Google is falling for these stupid tricks. But a lot of times when I get those reports, when I look into that, I see that we're ignoring that trick and they're doing five other things really well. And that's why they're ranking. So that's why I tell people not to focus so much on those particular tactics, but to look at your own website. Because those five things are doing really well, those are things you could be doing well too. But it's always frustrating. When you see some competitor doing something crazy and you're like, should I also do something crazy to kind of catch up? Or what should I be doing? Cool. Thank you. All right, let's see. Question that was submitted. I have a bunch of questions submitted. I don't think we're going to get to most of them. But OK, that happens. Question about a mobile site. If a website has part of a site responsive and part of an MDoc, I think we actually did this one. Good. Is having the same navigation appear twice in the cache version of a page OK? For example, a responsive site might have top navigation repeated twice in the cache version, because one is that stuff and one is for mobile. Does this affect crawl budget or search performance? For disturbing, there's somebody waiting. He wants to come and try you. But OK? OK. Um, I guess so. Yeah, do you know what his name is? Not really. He saw you at and wanted to join. OK. If you say no, I tell him no. OK. Well, more rush. Surprise test. OK. All right, kind of late for new people to jump in. OK, but fine. All right, so back to the navigation. That's perfectly OK. So with responsive sites, you sometimes have that, that you have different blocks that are sometimes visible for mobile, sometimes visible on desktop. And that's perfectly fine. So that's not something I would worry about. Would you say it's valuable that the title of a news article is visible above the fold? My theory is that it's more user friendly, that the reader sees it right away. I'm struggling to convince my team to switch the layout. I think that makes sense. So in general, when a user clicks on a result from search, I think it makes sense for them to get a confirmation that they're on the right place. And that usually is noticeable through title on the page, maybe through a header image or something on top of the page, or just generally the content of the page. I think that makes a lot of sense. So I would certainly try to at least get a title in there. It feels kind of extreme not to even have the title visible above the fold. Some sites are using meta refresh after five seconds and redirecting the user to a payment page for the content. In this case, does that impact their ranking? I still see the pages index with content behind payment and Google user can't see the content. What is Google's recommendation here? That sounds like a really bad practice. And I assume most people who visit your website once when you do this, they won't come back. So that's at least one group of people who are going to be upset. With regards to the meta refresh, we do see this as a redirect as well. So if you do this across your pages, there's a big chance we'll follow this redirect and think, oh, this payment page is actually what you want to have indexed, not the actual content. And in that case, we won't have the content index. So that's generally a bad thing, too. I have a website with more than 800 words and have less authority than my competitor. But my competitor's website has good authority and they have less content. So they're ranking higher than I am. Why might that be happening? So in general, I wouldn't focus on just the word count and just taking the number of words on a page as something to kind of give a sign of what this page is about. I'd really try to find kind of what is useful for a website overall. Hi. Nice to meet you. Great. Cool. Have more people can join in. All right. Happy to be back there? Perfect. So I wouldn't focus on just the word count. And that's probably not what will be driving things for your website. All right. I don't know. Any other questions from you all? I can see a whole day just in question. OK. Maybe for a preview, since you just dropped it. No, at the moment. Nothing. OK. Cool. John, can I ask something? That's OK. All right. Mihai, go for it. Well, it's regarding an email I sent you about drop site with that query there. This should be their main keyword that they should be ranking on. They're like page 3, 4, 5, or something like that. I know you mentioned you sent that to the teams. Take a look to get any responses back. I don't know which site you mean specifically. But maybe you can send me an email if that was something specific. It's a Rob's US site for experience gifts. OK. Right. And since January, they went from position 5, 6 to page 5, 6 all of a sudden. And that was pretty weird. I don't remember off end. I need to dig. OK. I'll just send you a follow-up then. Cool. OK. Yes, John, actually I have a question. All right. Yes, so actually for my site, it's our site, we have some pages are normal. It means Dextra, Mani, and some pages are in AMP. Many of the profiles, hospitals, and clinics, these type of pages in AMP. On May 19, from May 19, I see there is an increase in direct traffic in AMP, maybe in 10x. And same amount of traffic is getting reduced from the organic. So I don't know why this is happening. I don't know. I suspect these are just normal organic search changes that happen over time. So these kind of shifts, they can happen from time to time where we update some algorithms or tweak some settings on our sites based on what we think is more relevant for users. So that kind of change can happen over time. It's not that we would have a specific date for the most part where we say, oh, on May 19, we will change this setting. And all websites have to change things slightly like this. It's really just that these are normal organic changes that happen over time. No, actually, I want to say that this is happening for AMP analytics. So when I see the analytics report in AMP from May 19, sudden increase in direct traffic. But for AMP, direct actually generally, it's organic. I don't know. It might also be that there's something happening on the analytic side specific to that traffic. But I don't know how they're doing that. Joan, I've got one question. OK. The question is about site speed and the page speed tool that we all use every day. So our whole ecosystem, including our website, ucraft.com, is hosted on the Google Cloud. And basically, when I do the page speed test, Google Cloud, sorry, page speed is telling, hey, decrease your time to first byte metrics. But basically, I have spoken to Cloud team, Google Cloud team. They are telling that they cannot decrease that time to first byte metrics. And they are telling that just because of the fact that Google Cloud is redirecting the traffic from one place to one place according to their load balancers. And it requires for Google Cloud to require 150 milliseconds, at least, in order to redirect the traffic to the correct place. But then page speed is telling, decrease that 200. So I just want to understand how to take care. Yeah. I think that's always tricky. And you sometimes have to take these reports with a grain of salt. For the speed update, we're looking at different kinds of metrics, including some that are kind of calculated, like the page speed insights or the Lighthouse reports, but also some that are based on what users actually experience. And if you're looking at things overall, and it's like this one particular metric is just the one that you want to tweak a little bit more, then my guess is you're probably on the good side, and it's not something that you really need to worry about. It's something that might make sense to look at in the long run. But if you can't change that metric at all and if you're talking about 150 milliseconds or 100 milliseconds, then that's a really small amount compared to what it takes to completely load and render a page, which might be multiple seconds. So I assume that kind of a difference wouldn't be something that you'd need to worry about. OK, thank you. Sure. You had another question, right? Or you just have a millionth of questions. I'll ask one here. It's not very technical, to be honest. What do you think about teaching kids SEO? What's your personal take about that, like in school, where things like that are part of their education? I think it's an interesting because it involves different kinds of topics as well. So what I've sometimes done is gone to middle schools or high schools to talk about search in general. And I found that always really insightful because you see how kids search. And it's very different than how I would search, where they type like whole sentences and questions in. And that part is interesting for me. And I think that also shows that they have a slightly different mindset. But what I also find really important is to teach them how search works with regards to things like keywords in general, a search engine might think a page is really relevant. But actually, that's just because those words on a page, it doesn't mean that it's the correct answer. So if you, I don't know. So we have some example sets of pages that we create for talking with kids. And we give them these pages. And we say, well, for this search, it's like capital of Switzerland. And then we have a page about Zurich, which mentions capital in Switzerland on the page as well. It's like, when someone asks, what is the capital of Switzerland? And Zurich is the only page that you have that has capital in Switzerland on it. Then it kind of opens their mind. They think, OK, just because something is in search doesn't mean it's correct. So I think that aspect is really important. And then taking it a step further and saying, well, if you want to make a page to rank for capital of Switzerland, how would you do that? I think also it's very insightful for them, even without any of the technical aspects, like how to make HTML. But kind of the understanding, if you have a set of pages, how would you rank them? And then taking that to, if you want to make some pages for your own, what would you want to watch out for? I think that's really, really useful. I think in school, it's probably a bit early to go into all of the technical stuff for the bigger group. But there are always kids who are very technically minded, who are interested in these things, who want to kind of do everything. But definitely in college or university, when people talk about marketing or online marketing, I think learning about SEO and the details of all of that, that makes a lot of sense. What we do have presence in schools is with security. There's a lot of push from Google to ensure that families and children and their parents' teachers feel protected online. And we do go to schools and we teach these groups how to protect yourself and your identity. So maybe not that much SEO, but more about basic health practices online. Can you look at this? I love it. I'm going to keep going. I think there's a lot involved with some of these things. And even when we're talking about search, another topic is always like, if you put something online and that's the only piece of content with your name on it, well, what do you think the chances are that someone will be able to find it? To really bring them kind of the understanding that if there's something that you're doing online, there's a chance that it will remain findable for a very long time. So kind of think twice before you do anything really stupid, at least with your own name attached to it. Regarding removing something from Google, you said it might be that it stays online for a very long time. How hard is it to get something away, for example, a page that has existed on a website and now sends a 404 before that it has been indexed? The search and results page still shows the old snippet. And despite the server sending the 404, it keeps showing up. What is the suggested procedure? We have a URL removal tool that you can use. You can submit that exact URL to us and say this page has now been removed. And usually, within a couple of days, we will take that out of the search results. It's sometimes tricky because there are sometimes many different URLs that have or had the same content. So if you remove one exact version and maybe there's a version with dub, dub, dub, or with a parameter or a different upper and lower case, then maybe that version will show up. So you still have to watch out and maybe remove those over time as well, if you see that they keep coming back. It's a lot harder, I found, when the content is on other people's websites. So if it's a news article with your name on it and you remove the original source and maybe you cleaned up the problem, and not all websites are interested in moving like all of the content on them. And that's something that is not something that we can really change from Google side. I think within Europe, we have the right to be forgotten, which helps with names, at least. But when it comes to company names or when it comes to more generic terms, that's very hard sometimes. To be particular about what happened here is a colleague of mine made a site and unfortunately got hacked. It's an architect site. And during the time of the hack, it was redirected to some online shop. And then they got rid of the hack. And during the hack, they had a three or three, so basically to direct the traffic. And this happened about 30 months ago. And still, if I looked at up last night, the SERP does still show the result with the website's name of the architect. And then right after that, by Shorts and by Shoots from disaster company, despite them also having done the removal in the search console apparently he told me last night, and also having used the removal tool and the page in question also is sending a 404. OK. Yeah, if you can, you can have a look. Yeah, we can take a quick look. Yeah, yeah. But that was... I think it's tricky when a site was hacked and we have the title that was different. Like, you don't want to remove the home page because the home page is very important. And if the title just shows a part of the hack, then that just takes a while to be reprocessed. What I've also seen is that you remove the hack and you think it's clean, but it's actually still there. So then we re-crawl, re-index those pages, and we find the hack again. We still show it in search. And you think Google is just taking more time to actually fix the problem when it's actually still there. And some of these hacks are very, very tricky. That's a very good point, yeah, absolutely. No. And does it make sense to put these respective URLs in a special sitemap with current updates? To re-crawl them a little bit faster? To re-crawl them and speed it up a little bit? Sure, you can do that, too. Yeah, so if it's something that was changed, then that makes sense. In general, I would just use the removal tool if you can remove the URLs completely. And we should be able to at least find the main URLs fairly quickly, too. Sometimes it's more a matter of if you're searching for something very explicit, like you search for the architect and then buy shoes, then maybe we'll know, oh, well, this combination was only on this page, so we will still show it. But a normal user, they would just search for the architect and they might not see it at all. So that's also worth keeping in mind. Is it something that a normal user really sees? Or is this just because I know that there was a problem and I can find it? Is it because of that? All right, maybe we can take one last question and then we can close things down a little bit. Who wants to go? The right way to use analytics in the PageSpeed Tool in the Lighthouse Index, I found that the best thing to do is basically have a local version running on the server and keep that updated instead of using the online version to not have issues of this. I don't know, the script has to be updated or it has to be cached, I think something like that. There's a warning message I'm not too sure. For Google Analytics? Yeah, for the script itself. And generally, when I do client sites, I like to get 100% in all metrics and make it really nice, a bit of a fanatic thing to do. And then some blog posts suggested on what's a local hosted, hosted on the server and keep it updated whenever Google pushes a new version. Put that on your server and you should be fine. I suspect if we ask the analytics team, they will say it's better to use the live version. But if you can really make sure that your local version is kept up to date, then that should be fine too. OK, thank you. It's sometimes a bit tricky with regards to the analytic scripts or maybe the AdSense scripts or scripts like that in general in that they often serve content that's optimized for specific browsers. So if you use a kind of a default testing tool, like page feed insights, like the live version to test that, then often you'll see this script is not optimized. But if you use it with a browser and you check what is actually sent to the user, then you see, oh, they're doing it right. So it's just that they're recognizing this tool that's accessing my page is not a normal browser. So we will send it kind of a simplified or a special version. And then you have that discrepancy. But if you really want to get those 100%, I don't know. If you really can make sure that the locally cash version works well, that it's kept up to date, I think that's fine. No. You would agree that having those 100% is important for its placement as well? Or is it such a huge factor or not? So that's one of the reasons why we're trying to get rid of those numbers, kind of those scores, because we see people focus so much on those scores. And then they don't realize that maybe there are other aspects of the page that are a little bit slow, but just because this one score is from Google. And we think I need to get 100 on everything from Google. Sometimes we see people focusing way too much on those details rather than on the bigger picture. But I think it's completely natural. If you're a competitive person and you see a score, then you want to get that score as high as you can. Now, that happens. All right, let's take a break here. I imagine we'll have more discussions a little bit. I'll let you all kind of keep in here as well, but I'll pause the recording so that we have something reasonable on YouTube as well. Thank you all for joining. It was great having you here. Nice to see some faces in person. Nice to see some of you again as well. And maybe we'll see each other again in one of the future Hangouts. All right, bye, everyone. Thank you. Goodbye.