SEO for Google News

Barry Adams explains how Google News indexing and ranking differ from regular search. He covers technical requirements like AMP and structured data, headline optimization, the importance of recency, and how the March 2018 update changed what Google rewards in news content.
Chapters
Hello, everybody. I'll try not to speak too fast to give the awesome translators a bit of time to catch up for those of you who don't speak English. I'm going to be talking about Google News. It's a bit of an interesting vertical within Google itself. It's been around for quite a long while. I've quite a lot of experience with Google News. I work for these publishers, including Fox News, which is the most popular cable TV channel in America, and the Daily Mail, which is one of the most popular English-language newspapers online. Although in the UK they were recently overtaken by The Sun, which is also a client of mine. So that gives for some interesting meetings sometimes. Google News is quite an interesting beast, slightly different from normal Google, although we do see the two merging more and more. But before I explain what Google News is and how it works in relation to regular Google and regular SEO, It first is important to understand what SEO itself is. I mean, we've been talking about SEO pretty much all day here, but do we really know what it is? Do we really understand what it actually is? What is it that search engines actually do, and how can we learn from that? Because what search engines do has a name. It's called information retrieval, and the founders of Google, Larry Page and Sergey Brin, studied information retrieval at Stanford University while they were building Google, then called BackRub, as a university project basically. And the book that they used to study information retrieval can be read online for free. You can go to the Stanford University website and read the Introduction to Information Retrieval book yourself. And if you're serious about doing SEO, I would recommend you try to at least read some of it. It goes into quite detailed mathematics quite quickly, at which case I'm totally lost, but I think just grasping some of the basic concepts of how search engines work and how information retrieval works can be very, very useful. One of the things I learned, for example, when I first read this book 10, 12, 15 years ago, I don't remember, was that every search engine has 3 main components. There's a crawler, an indexer, and a ranker. And in Bartosz's talk, he touched upon this as well. The different processes do different things with the content that it finds on your website. And this is going to become important when we talk about Google News, because Google News doesn't necessarily touch all of the aspects of the search engine, because news is very fast-moving, and therefore Google doesn't necessarily have the time to spend, you know, really indexing content in depth. It has to be very quick at how it handles news articles. So with this process in mind, these 3 main aspects of a search engine— this, by the way, is how Google itself maps it out in their own little diagram. The crawling system is quite advanced, a lot of stuff going on in there, and then there's an indexing system which is that 2-stage process that Bartosz talked about. First, raw HTML, second, executing all the JavaScript, and then the ranking engine which is the front end of Google, is the green box there at the end. When I do SEO, I have a 3-pillars of SEO model: technology, technical SEO, relevancy, on-page SEO, and authority, links and link building, what Kevin was just talking about. And this of course relates to the 3 main processes of a search engine, where technical SEO is about the crawler, relevancy is about the indexer, making sure the content can be understood, and authority is about the query engine, or basically about ranking first in Google if you can at all. So that's the model that I use to approach SEO, and Google News fits in that model as well. And you will, you will see that Google News is slightly different from regular Google and does things slightly differently because it is so fast-moving, and Google doesn't necessarily have time to really go in-depth and crawl a website over and over again. So we have to give Google some extra tools as a news publisher to allow your articles and your news content to be very quickly crawled, indexed, and ranked. So Google News itself has been around for quite a while. It started as a pet project back in 2002, and I'm old enough to remember that. Then in 2006, it was officially out of beta, which basically means that Google said, right, we're done tinkering with it, we can actually publish this as a product. In 2007, it was integrated into Google's regular search results. So you had the universal search box in Google with top stories in regular Google Search. And in 2015, there was another big shakeup when Google introduced Accelerated Mobile Pages, which we're going to talk about a little bit as well. And I'm not a big fan of AMP, but we sort of have to talk about it a bit. And then earlier this year, we had another really big shakeup when Google revamped Google News entirely. And now they're saying that new Google News is powered by AI, by machine learning and machine intelligence. Which is a really easy way for Google to say, whatever you see in Google News, it's not our fault, it was the machines that did it. And that really pisses me off, because who coded the machines? Exactly, the Google engineers. It's just a way for them to hide behind the machine system and say, if you read bad stuff in Google News, it's not our fault, it's the machines that did that. Which is total bullshit, of course. This is what Google News looks like, at least in the UK. Fairly clean layout, but not a lot of people actually go here. If you look at the traffic that Google News, that particular vertical, gets— google.co.uk or google.ar/news— it is 1.8 to 2% on average of most news publishers' traffic. That's not a lot of traffic. So why is Google News such a big deal? Well, Look at that graph. That is Google Search in general. Half of news websites' traffic generally tends to come from Google Search. But that's not Google News. Or is it? Well, yes, it actually is, because that is that— the top stories carousel, which you see on roughly 11.5% of all Google search results. Now, Google does— what was it— 20 trillion searches every year. So we can say that 2 to 3 trillion of those have a Google News Top Stories box in there. And yes, you have to be in Google News, more or less, to be in that Top Stories box. They appear for all kinds of different searches, no matter what you type in. It doesn't have to be a current affairs topic. It can be something else. It can be about currency prices. It can be about restaurants. It can be about anything where there is a publisher that writes content about that. That is included in Google News, and you will get a top stories box. And yes, you have to be included in Google News as a publisher to be in that box. The pink/purple bars there are websites that are in Google News that appear in one of those news boxes, and the yellow ones are websites that appear in the news boxes that are not in Google News. As you can see, Google News included websites tend to dominate those boxes on desktop search. Sometimes you see articles ranking in there from websites that are not included in Google News, in the manual Google News index, but it's very, very rare. Of course, we can't ignore mobile. Mobile is the biggest form of search at the moment. At least 60% of searches happening on Google happen on a mobile device, smartphones and tablets. And of course, the same applies. We have those mobile news carousels on mobile search as well, and they're very prominent. You can scroll through them, swipe through them, and see the news articles. And you also have publisher-specific carousels. If, for example, there's a news publisher who writes a lot about a specific topic and you search for that topic, you might get a carousel like that a bit lower in mobile search results with articles on that topic from just one specific publisher. This, for example, is in the carousel for the Brexit keyword, and The Independent has its own publisher carousel because they write a lot about that specific topic. And yes, here too, being included in Google News is pretty much mandatory. Although, as you can see from the size of the yellow charts, there is a bit more leeway in mobile search results. Google is a bit more inclined to include non-news websites in the Top Stories carousel on mobile. Versus desktop. But generally speaking, most websites in the top stories carousel on mobile will also be Google News included websites. So that top stories carousel sends a lot of traffic to news websites, and that is where they get the majority of their search traffic from. And you will see from the little AMP icon there in the, in the mobile carousel that yes, you have to have an AMP version of your articles to show up in that mobile carousel. I think it's like 95% of articles in the mobile carousel are AMP articles. So Accelerated Mobile Pages is what they are. It's a Google pet project that I am not a fan of. They're basically trying to reimagine the web in their own image, making it nice and clean and lean. And as Bartosz had shown in his talk, AMP isn't necessarily faster. Than regular web pages, but Google cheats a little bit. They do some pre-rendering and caching on their end in the AMP cache and making it appear faster for users who are on their mobile devices, especially if you're on an Android device. So AMP basically means you have an alternative version of your articles which is built specifically for mobile. It's a fairly restrictive standard. Not everything that you can do on a normal article can you also do on an AMP page, especially when it comes to advertising and tracking. And it is an all-or-nothing standard, so either you have a valid AMP page and you can appear in the mobile carousel, or it's invalid and Google will reject it and you will not appear in the mobile carousel. So you sort of have to play by Google's rules. If you're a news publisher and you say, you know what, screw that, I'm not going to be using AMP, you will be missing out on a lot of potential traffic. You will not get those 60% of mobile searches, And you will not get the vast amount of search traffic that most news publishers get. So you have to sort of bend over and play by Google's rules whether you want to or not. This is a graph that demonstrates that very clearly. The Belfast Telegraph is the red line. They decided not to implement AMP because their owners are based in Dublin. And in Dublin, in google.ie, they were very late to adopt AMP. Whereas the Irish Times thought, you know what, we are going to implement AMP. And they were really quickly gaining on the Belfast Telegraph in visibility in mobile search results off the back of that. And you can see exactly when the Belfast Telegraph decided to actually implement AMP, because they started to reclaim some of that lost visibility. So how do you actually get into Google News? Now, Google News is a manually curated index, which of course is very different from regular Google web search. Regular Google search, you don't have to submit your website to Google. It'll just find it, hopefully, crawl all your links and links pointing to your website and find all your content that way. For Google News, that is not the case. Google News, you have to submit your website to the index and hope you get approved, and then you will be included as a news publisher in Google News, and you might get all those benefits. You might get in the top stories carousel, and your articles might get an awful lot of search traffic. So how do you do that? Well, it's a fairly extensive process. You have to be first and foremost a proper news website. I get this question a lot from potential clients asking me, right, we have an e-commerce website, or we have a gadget website, and we also have a section about news in the industry. Can we submit that section to Google News? The answer is no. Well, You can submit it, but you'll not be included. They will just tell you to get out. You have to have a dedicated news website, a website whose primary purpose is news. And that news can be on any given topic. It doesn't have to be general news. In fact, it nowadays helps to have a very specific editorial specialty where you write about a specific niche topic, a specific industry. 'Cause Google is very keen on including more websites that are very niche and very focused on smaller topical areas. General news websites, we sort of have enough of them already, so Google isn't really looking at including more of those. You have to write a lot of original content as well. You can have some syndication, some republishing of articles from like wire services like AP or Reuters, but generally speaking, Google wants you to write most of your content yourself, and ideally multiple articles per day if you can, because just one article a day is not really a news website, is it? Having multiple contributors also helps because it makes you seem more authentic and more like a proper news organization. And there are some technical requirements as well, like having a news sitemap and accelerated mobile pages. Now, Google says both of those are optional. In my book, they're not really optional. You saw the have to have them. And the news article structured data, that is definitely a must-have, otherwise you will never get in. Google needs to see that structured data on your article pages to recognize them as articles, and therefore allows them to index that very, very quickly. And Google is working to clean up its own index. They kicked out 200 publishers in the early days of 2017. Since then, they've kicked out several hundreds more. Websites that are trying to play games with the Google algorithms and trying to trick Google into ranking them get kicked out. Websites that aren't publishing real news, that are more fake news websites, they get kicked out as well. Google is working hard to keep their news index clean if they can. So once you're actually in the index, if you have a proper news website and you've jumped through all of those hoops, how do you get to rank in Google News? How do you claim that traffic? How do you get that top stories carousel to feature your articles and not those of your competitors? The editorial specialty of your website is probably the single biggest factor there. If you write about a specific topic very often, Google will recognize that you have that editorial specialty and will see you as a trustworthy news website to publish on that content. So for example, Search Engine Land is one of the most reputable websites when it comes to SEO. Any given SEO topic, Search Engine Land will be featured in the top stories carousel on mobile and desktop because they are seen as that reputable source that writes good content on the SEO industry. The same goes for any given other topical news. If you write something about— if you search for something about, for example, exchange rates between different currencies, you will see websites in that carousel, which are primarily financial news websites or exchange rates-focused websites. The other one is, of course, the headline. Because Google News has to work so fast and try and index articles very quickly to show in the Top Stories carousel, it doesn't do an in-depth analysis of the content. It basically looks at the headline and title tag and sees, right, what is this article actually about? So you have to put the keyword or the topic you want to rank for in the actual headline. And I've used this example on purpose because I do like my whiskey. And it's not just about having the keyword in the headline, it's also about putting it towards the start of the headline. Google News puts a lot more emphasis on the first 6 words of the headline and sees what keywords are in there. So if you want to rank for a topic, like be it a football player, a politician, a brand, whatever it is, Make sure that the article has that name in the key— in the title, in the headline early on, in the first 6 words if at all possible. Also, try to avoid jokes or puns. You know, there's a lot of journalists who like to write headlines that are a bit funny, a bit of a play on words. That shit does not work in Google News. Google does not have a sense of humor. So don't try to be funny. Just stick to the facts. Thanks. Put a factual headline in there if you can. The headline also has a fairly strict character limit. The news article structured data that you have to have implemented on, on your article pages has a hard limit in the headline field of 110 characters. Now there's a lot of news publishers who like to write longer, more elaborate headlines to try and make the story come to life a little bit, and they see that as part of their brand. And they will run into this problem quite often. If the headline is 111 characters long, it'll break the structured data, and Google will just not see that as a news article and will not include it in the Top Stories carousel. So you have to implement that hard limit of 110 characters in your headline. Recency, of course, is a big thing. Google prefers content that was published fairly recently or updated fairly recently, And you can play around with this a little bit. For example, you can publish an article on a topic that is currently developing with only like 100 words in it, and then you later update it and you expand it a little bit and you give it a new publication date to reflect that, and Google will see that as, okay, maybe we need to re-rank this again. The most extreme version of this is when you just republish the article with a new URL, which basically means Google will see it as an entirely new article rather than an update of your existing article. And that can be a very clever way to artificially boost your rankings in Top Stories by republishing it, tweaking the content a bit so that it is a bit different, and maybe changing the headline a little bit as well, and then giving it a new URL so Google sees it as a new article. You do that in the course of the day several times, and you will find yourself back in that Top Stories carousel Fairly quickly. Now, don't abuse that too much, because I have seen websites get kicked out of Google News for doing that too often. I think it works best if you do that maybe once or twice a week, and only for topics that you feel really confident about, that you are actually generating newsworthy content about. Don't just do it to get cheap traffic, because eventually somebody at Google News is going to notice and going to be pressing that really big, fat, red Delete button. And off you go. There's a lot of other factors that go into it. Like I said, the structured data, the headline, make sure the headline is highly visible in the HTML source code as well. Google says there's a minimum limit of about 80 words for an article. Now, I have seen Google reject articles as being too short that are up to 200 words long. So I think it depends on more than just the length of the actual article. And the article fragmentation is a bit of a problem. And you will see from that screenshot that is from the old Google Search Console, we used to have a report there, a crawl errors report, specifically for news websites. If you were included in Google News, you would get a news tab in your crawl errors report in Google Search Console. And that would give you all kinds of really, really useful information about how Google would be crawling and extracting the articles from your web pages, and the kind of errors that you would get would determine what you needed to do to make sure those errors decreased. Of course, Google being Google said, you know what, fuck you, we're gonna take that away. Thank you, Google. The single most useful report for me as a technical SEO working with news websites, and Google said, nah, you don't get that anymore. Thank you. So what do we do now? Well, we can take some learnings because fortunately I spent a lot of time analyzing that report before Google pressed the big fat delete button, and we have a few things that we can learn from that. For example, one of the common errors I tended to come across was title not found. Google would try to index the article but couldn't find the title of it. I'm like, why? That's a perfectly good title. You know, why does Google not see that? Why does Google not recognize that this article just has a normal title there? Well, the answer is in the source code of the actual article. Because there are some special characters in there that break Google's indexing of the article. So having special characters in your title breaks Google's indexing. They can't extract the title, so it thinks you don't have a title there at all. Just by having these special characters in there, in this case little quote marks that get turned into ASCII code in the source code, breaks Google News' indexing of those articles. Little things like that can be very, very powerful and very, very dangerous. Or take this one: article fragmented. What does Google mean by a fragmented article? The article's fine, it's just, you know, normal stuff. Why wouldn't Google News be able to understand this? Again, The answer is in the source code of the article. When you look at the source code of the article, there's an introduction paragraph, a section with some recommended news, and then the rest of the article. And it's that middle section that breaks indexing. That middle section interrupts Google's processing of the article and causes Google News to just reject the article as being fragmented. It can't understand the actual topic, the actual length of the article, what is part of the article and what is not. Part of the article. And more and more of these errors I saw, the more and more I realized that the indexing process that Google News uses is just the crawler itself. The crawler does the heavy lifting when it comes to Google News, not the actual indexing engine. The crawler does first-pass indexing. If you were at Bartosz's talk, he talked about the 2 stages of indexing. The first stage is based on HTML. The second stage also executes the JavaScript. Google News only uses first-stage indexing. It only looks at the raw HTML source code and doesn't do anything with it. It just parses it and throws it into its index straight away, which is why that first-stage indexing process is nearly instantaneous. It's really, really fast. And that's what Google News needs. Google News needs a fast indexing process because news moves very fast, articles get published all the time, new stories develop, so Google needs to focus on speed. And the crawler has that process built in, which is why Googlebot— when Google engineers talk about indexing, they tend to talk about Googlebot, even though Googlebot is technically the crawler, not the indexer. But Googlebot, the crawler, does the first stage of indexing based on the HTML source code. So when we publish articles and we want to make sure they're included in Google News, we need to make sure that the HTML source code is clean and tight and lean and doesn't have a lot of junk in there. So a nice clean set of code from the headline all the way to the last paragraph that contains the article so that the HTML parser that is built into Googlebot can very quickly extract it and index it and doesn't have any errors associated with that. This article I wrote a while ago for State of Digital has a lot more of those common crawl errors explained, so I would recommend you read through that and read through all the other stuff that Google does when it extracts articles for Google News. But being in Google News is just the first step. As a publisher, being in Google News, you get more than just the top stories carousel. You can leverage your, your strength as a news publisher to win in regular search results as well. For example, any given topic you type into Google, in the regular results you will see news websites in there. You will see the category pages or the topic pages for that particular search pop up very high in Google search results. Why is that? Why do news websites dominate regular Google so much? Well, the answer is in what Kevin was talking about just before me. It's all about links. All these news publishers have fantastically powerful link profiles because there's an awful lot of links pointing to those websites. These are screenshots from Majestic. The Trust Flow and Citation Flow metrics are a really good way to indicate the strength of a website's link profile and how many links and how high quality those links are. And the process goes a bit like this, where if you're a news publisher, you publish news articles, you tend to get links pointing to those news articles because you're seen as a news publisher, as a trusted source. So people will cite you and report on what you're reporting and link back to you, which means, you know, those articles will start to rank very high in Google. And that means you're seen as an even more trustworthy source, so it becomes a self-reinforcing cycle that generates an awful lot of links. Which is one of the reasons that I, as an SEO, love working with news websites. Because link building— I hate link building because for link building you need to have social skills, you need to like people. I don't. I hate people, just give me computers. But news publishers, link building is not an issue. It's not about, oh, we need to get more links to the site. No, it's more like, what the hell are we doing with all these links? We've got 10 million links pointing to our site. Yay! So it's all about that process, about leveraging that link power. How do you get— because the article pages are the ones that win the links. It's not the category pages. The article pages get links. So then you have to look at leveraging those links to make your category pages rank high. So as an example for José Mourinho, you search for that in Google UK, you get The Sun and The Guardian. Category pages on José Mourinho ranked very, very high in search results. And why is that? Those pages themselves don't have a lot of external links pointing to them, but the articles that write about them have a lot of links pointing to them. These are the 2 category pages, and these are the articles that write about José Mourinho. And you see there on The left-hand side, the links from the article itself directly to that category page. And on the right-hand side, the tags, the article tags again pointing to that category page. So what is happening here is the articles get all the links from external sources. You know, somebody writes something about José Mourinho in The Sun or in The Guardian newspaper, some sports blogger picks that up and writes about it as well and then links back to the original source. So the article on the Sun now has incoming link value. That link value is sent through those internal links to the category page so that the category page also gets a section, a slice of that link value, and that helps the category page rank high. And this is a process that is continuously ongoing, which is why those internal links are so incredibly powerful, because the more articles news websites write about a specific topic and the more they use those internal links, the more the category pages are seen as highly trusted sources and will rank high in search results. The Daily Mail is very, very good at that. They sort of pioneered this process, but they went a little bit too far with it. What Kevin also alluded to in his talk, you have to be a little bit careful with this. You don't want to overdo that too much. These are 2 articles on the same topic. One is from the Daily Mail, one is from The Sun. Both are my clients. Ironically, the Daily Mail article has 17 internal links to different topic pages, whereas The Sun article only has 4. And the Daily Mail uses that very aggressive internal linking, and they used to be by far the biggest news publisher in the UK and the largest news publisher in the English-speaking world. Not so much anymore. The Sun used to be behind a paywall. You want me to go back a slide so you can take a photo? The Sun used to be behind a paywall, and of course that paywall was absolutely catastrophic for their search visibility. You throw up a paywall, Google can't crawl your website anymore, no more visibility in search. So they dropped the paywall, and I started working with them shortly thereafter. We started building a new website, and we had really, really good growth trying to, you know, overtake our old position in search traffic and then trying to grow beyond that. And of course we started gaining on the Daily Mail. One of the biggest reasons we gained on the Daily Mail was because The Sun was smarter with their internal linking. Daily Mail puts 15 to 20 internal links in an article, whereas The Sun only puts 2 to 5 internal links in there. You think, well, more links is better, right? It means more link value is sent to those pages. Well, not really, because Google has gotten smarter about that internal linking and about assigning the right value to those internal links. And what I think actually happened is that because of the Daily Mail's very aggressive linking strategy, they were seen as more of a publisher of thin content, content whose primary purpose was SEO and implementing things to improve rankings rather than content that actually served a news purpose. The difference is also the Daily Mail publishes around 1,000 to 1,600 articles per day, whereas The Sun only publishes around 500 to 600 a day, which is still a lot, but not as much as the Daily Mail. But The Sun was winning this battle, and it has actually won. The Sun gets now more traffic from search in the UK than the Daily Mail does. And I think it's because the content that is published by the Daily Mail with that very aggressive internal linking is devalued by Google, and the links in those articles are not really carrying any weight anymore. Because Google realized the Daily Mail was being too aggressive and trying to use internal links to manipulate rankings. And Google doesn't like it when you do that. So you have to play a little bit smarter. You have to tone it down a bit and be very smart with your internal linking and not be too aggressive and too over-optimized. A few internal links, great, but don't throw 10, 15, 20 links in there because Google is just going to say, uh-uh, you're not going to get the benefit from that anymore. And then of course we had something very special happen in March of this year. The 9th of March is what they call News Apocalypse. Because something happened that affected news websites in the UK and only news websites. This happened. Blue is the Daily Mail, red is The Sun, then you have the Express and the Mirror. Basically all the tabloid news websites in the UK were absolutely slaughtered on March 9th, 2018. They lost roughly half their traffic. The Daily Mail lost half their traffic, The Sun lost about a third of their traffic. And other websites were similarly affected. So we were like, what the fuck? What? What? What? Don't you like us anymore, Google? What happened? Google actually admitted that they rolled out a core algorithm update, and they said as well, there's nothing you can do to fix it. Ha ha ha ha! Gee, thanks. And as we started to analyze this update, we we sort of found out what Google actually did. The old method here where, you know, you would get links to your articles, would build a lot of authority to category pages, etc., etc., you become a top publisher, you can rank for nearly any topic as a news publisher. That wasn't entirely true anymore. It still applies to a large extent after March 9th, but with the caveat that as a news publisher, you have to stay within your topical focus. You have to only write about what you're historically good at in terms of your journalism. I call it the stay-in-your-lane update. I think the official name is the Fred update, which is a terrible name. I call it the stay-in-your-lane update because what basically happened here is that Google thinks that as a publisher or as a website in general, you have a certain topic, you have a certain set of topics that you're good at and that you can rank for. And if you write content that is outside of that topic, no matter how fucking good it is, you're like, nah, not gonna rank that because it's outside of your lane. You have to stay in your topical lane. And this can be very clearly demonstrated with this example. Up till the March 9th update, The Telegraph, a general news website in the UK, ranked in the top 1, top 2, top 3 for top 100 songs of all time. They would— they had a good article about a long evergreen article, and that is a very interesting search term because a lot of volume for that on a daily basis. And you could— they would get so much bloody traffic from that. After this update rolled out, this entire search result dramatically changed, and it's now mainly music websites that rank for that because Google sees that search as a music topical search, of course it is, and thinks, right, no matter which publisher writes about that, unless you're a news— sorry, a music-focused publisher, you're not going to rank for that. Only music-focused websites can rank for music-focused searches. And The Telegraph, being a general news website, doesn't focus on music very often, so they're not getting that result anymore. I think at the bottom of page 1 at the moment, just above the fold, or maybe just on page 2. They're not getting that massive amounts of traffic anymore. So that was a very big game changer for the news industry, and realized that, right, as a news publisher, you can't just get free traffic to any given topic anymore, which used to be the case. It was fantastic, those old days of news. You could write about any given topic and you would rank near the top of Google because your link profile was so incredibly strong. Now it's more about, are you specializing in specific topics? Do you have that editorial specialty, that topical focus on these certain areas? And Google will allow you to rank within those areas, and outside of those areas, you don't get that traffic anymore. So what's coming up? What is in the future for search? What other things will be affecting how we're optimizing news websites for Google and the kind of traffic we can get? Well, you will have heard this as a common theme in the last day and a half as well. Voice search is going to be more and more important, and the speakable structured data is going to be playing a very big role with that. The speakable markup is basically structured data you can implement on your article pages that indicates, like, these certain sections on my website can be read out loud by a voice assistant like Google Home or Amazon Echo or whatever you have. So this tells the search engine you can read out the headline and the summary in a voice assistant when someone wants you to read out the news to them. This is incredibly powerful stuff if you implement this, and Google rushed this through the spec and has started adopting it even though I don't think it's officially approved as a schema.org specification yet, but Google already supports this because it is so incredibly powerful. This allows Google to tap into voice search, which is a battle they're losing at the moment because Amazon is winning the voice search battle. Amazon with the Amazon Echo and the Dot and all those things, and the Alexa, is beating Google at this game. So Google wants to become more useful than Alexa, which is why they rushed this through and are encouraging publishers to implement this speakable structured data so that, you know, their news can be used by the Google Assistant and read out loud to people who are like making breakfast or whatever it is, or sitting in the car, and the news can be read out to them. Another thing is again structured data. The news article markup is being expanded where there's more specific markups for specific types of article. So if you have an analysis on the current affairs, you get slightly different markup than if it is an opinion piece or if it's a review. And this allows Google to more accurately interpret what you're writing about and when that content should be shown in search results. And this again will be very powerful. This is still in the pending stage on schema.org, but again, Google is very enthusiastic about this and wants news websites to start adopting this. So if you do work in a news publishing environment, Do keep an eye on this and be ready to adopt this fairly quickly when Google says that they're going to support it. We also have something called NewsGuard, which launched in August. You may have noticed in some of the screenshots I showed you, there were little blue tick marks there. This is the NewsGuard plugin. NewsGuard is a project started by a group of journalists in, in the US where they will manually evaluate websites to see if they adhere to journalistic best practices. Do they cite their sources? Do they write corrections? Do they have a proper fact-checking process? Etc., etc. This is a manual process whereby journalists working for NewsGuard will phone up these news publishers and contact them and look at the process that they use to update news and to publish news. And if they adhere to journalistic standards, you get a nice green mark. If it's more hit and miss, you get an orange mark. And if they think you're a bad news publisher that does not adhere to proper journalism, you get a red mark, which is quite problematic. Now, NewsGuard is only a plugin at the moment that you can install in your browser, and you get these highlights in Google Search as well as on Twitter and on Reddit when you see articles linked there. But I think it's a fairly clever mechanism because once they've gone through most of the world's news websites, The effort becomes significantly less. They don't have to spend so much time evaluating new websites. They can just periodically review new websites to see if they've improved their standards or not. And because Google is fairly focused on combating fake news and trying to clean up the top stories carousel and make sure they only publish reputable sources, it wouldn't surprise me if Google were to say, you know what, we're going to start collaborating with NewsGuard. And maybe do something similar. Because Google have their own news initiative. The Google News initiative is Google's way of trying to promote proper journalism on the web and not reward fake journalism and keep journalism. They're trying to make sure that traffic goes to the right news websites and not the wrong news websites, whatever right and wrong might mean. It's worth keeping an eye on that, subscribing to the newsletter, and see what they're up to, because they're also giving a lot of money away to small publishers and small local organizations. And maybe you might apply for one of those grants as well. So in a nutshell, what is Google News and ranking in Google News about? It's fairly simple. You know, once you're in the Google News index, you can get a lot of free traffic, provided you write about your specific topic. You can leverage your internal link power to drive a lot of traffic outside of the news cycle to your category pages. You have to make sure you have these technical things like structured data and the XML sitemap implemented, and you have to stick to your topical expertise. It's very hard to rank outside of your topical expertise as a news publisher nowadays. My name is Barry Adams. I've been doing SEO since way too long. If my talk wasn't impressive enough, I hope those logos will impress you. Thank you very much.
More from Barry

NewzDash Webinars: Barry Adams ON Advanced Tech SEO for News Publishers and LIVE Q&A









