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

Hello everyone, welcome to another episode of NewsDash Webinars. Uh, we have amazing, uh, Barry Adams, a good friend of mine and a brilliant SEO and, uh, very advanced tech SEO for news publishers as well. Barry has been Doing SEO since what, 1998, 1999?

Before it even was called SEO, I knew what it actually was professionally, about 2004, 2005, I'd say.

Yeah, so Barry has been doing it for a long time. I always like to joke and say, hey, we have been doing SEO since Google Dance, or a little bit before that. So yeah, for the veterans out there, um, Barry is also the co-founder of Ness Um, and, uh, he is the, uh, I don't know, the author, the owner, uh, and the brilliant mind behind seoforgooglenews.com as well. Uh, so please welcome, uh, join me welcoming Barry. Barry, it's great to have you today.

Thanks, John. That's quite the introduction. I'm blushing a little bit, to be entirely honest. So, uh, I hope I can live up to the expectations.

No, I think, I think a lot of people have great anticipation for this uh, webinar, I had— me personally, I, I put like, okay, this is every question I get to ask Barry that I never asked before. So I have a list of questions, uh, but today we will, um, split our session. So about maybe 45 minutes, uh, we're gonna be focused on some of my questions, uh, and then we're gonna have a short, uh, presentation by NewsDash for 2 minutes, and then we're gonna continue with Q&A for like your own questions. So if you have any questions that you want to ask Barry for the, for the second session, please on your chat on the right side panel, if you click Q&A, please add your questions there. Don't add your questions in the chat, and then we will answer all of them. So Barry, welcome again, and let's start with some of the general questions. What do you think distinguishes news SEO from traditional SEO practices, right? Why is it important right now to— like, for the longest time I have been trying to establish news SEO as, as a branch of SEO, and I think the industry has matured enough and we see a lot of news publishers have and hire either news SEOs or SEO editors and stuff like this.

What, in your opinion, what distinguishes Um, news has an interesting role in Google's ecosystem cuz it's, it's one of the few areas of, of content where Google doesn't have a lot of time to carefully consider it. Um, because the speed with which news moves means that Google has to crawl, index, and rank news articles very quickly within a few minutes. Um, Google is capable of doing that. Google doesn't like doing that because it means they have to take shortcuts. Yeah. They have to take shortcuts, especially when it comes to crawling and indexing, uh, to make sure that the, the articles rank really, really well. Plus, you know, people like to complain about Google search results and how much spam is in there. Um, and, and that would be, uh, even worse if that happened in the news-specific boxes, which is something Google has actually had issues with in the past. Yeah. So Google has to rely very heavily on trust and authority signals in its news ecosystem because expectation of quality. And when it comes to seeing news stories in Google search results, where if it's like a regular search result, a regular 10 blue links, you, Google can sort of get away with not always showing the, the most authoritative and most trusted, uh, sources. Whereas in news, Google doesn't have that luxury. It has to rely on, on strong authority and trust signals, much stronger than in, in regular SEO. So good, like news is definitely very different from, from regular SEO in that it's much more It's much harder to get traction as a new website in Google's news ecosystem because you have to build those authority signals and you have to focus a lot more on how quickly and efficiently your site can be crawled and indexed so that newly published articles can, can rank as, as fast as possible whenever they're published.

Yeah, uh, absolutely. And then would you agree with me, news SEO is more satisfying as well versus SEO? Like in SEO, you do all the optimization and the work and stuff like this, and then you hope in 3 months, to show to your boss. Uh, but with news SEO, if you have that authority and you do the, like the news SEO the way it should be, you can see results in minutes.

Oh yeah. It's, it's the closest to real-time SEO as you can get. Uh, when an article is, is published, Google will crawl and index it usually within a few minutes. Um, at which stage that is generally your one chance at getting search traffic to that article unless it's an evergreen piece. But if it's a news story, you, you basically get one chance at it. The shelf life of an article in Google's news ecosystem is maximum 48 hours. Generally a lot less than that, 24, 32 hours if you're lucky. Um, and when the article initially starts to rank, that version of it, the one that Google sees initially, is generally the version that will stay in Google's index until, you know, Google refreshes its index, which can be several days later. At which stage it's not a news article anymore. So you get one chance at it, which also means that the moment you start doing things slightly differently with your news articles, like maybe different types of headlines, different imagery, um, new meta tags, whatever it may be, you tend to see those results very quickly, um, because of the real-time aspect. You know, you don't have to run a, a, a, a separate test like you have to do in classic SEO for 3 months before you have an, an impact. a bit of confidence in what the result will be. With news, within a few days or within a week, you tend to know if something is working or not. Yeah.

So over, over this 2 decades, uh, uh, or more of, of SEO and news SEO, what do you think has changed, especially over the last few years? Right. So I think 2 years, uh, 20 years ago, it was so easy, right? You put some keywords on the page, hide some, you know, do a few things and you're done. Right. You own the SERPs forever. But it's, it has been much harder, right? And a lot of SEOs are asking, uh, what's the future of SEO? But before we get into the future of SEO, I wanna see like, what do you think over the last few years, what has changed with SEO and news SEO in particular?

Yeah.

I mean, we have to keep in mind that Google, when they launched Google News as a separate vertical, uh, it had its own crawler and its own index.

Yeah.

And that was by necessity because at that time Google's regular crawling and indexing system was quite slow. It could take weeks for new articles to start showing up in the index after Google had initially crawled them. Um, and that of course didn't work for news. So for Google News, they had a separate Googlebot News crawler and a separate index, which was also a manually curated index. You had to be approved to be included in that news index. And that crawling and indexing process was, um, fairly straightforward, I would say, because it was very much focused on speed and not, not on doing a lot of, of sanity checking, deduplication, and controlling again, because it needed to be fast. Um, and that's also why it was a manually curated index. So Google would only let websites that it knew were reliable publishers into that particular separate Google News ecosystem. That started to change from like 2012 onwards when Google's regular crawling and indexing system became fast enough and good enough that Google didn't need to have a separate index. And I think about 2018, 2019, Google completed that merger. And now everything in the news ecosystem is running off the exact same infrastructure with the same crawler, same indexing system, same ranking systems as, uh, regular Google. Now, the thing that have, has really changed is the algorithm updates, not just for news, of course, but also for all SEO that, that news publishers, because they are now part of Google's regular crawling and indexing and ranking, uh, infrastructure, are now also subject to the ranking algorithm changes that Google makes to protect all their rankings, not just their news-specific rankings, because, uh, Google now uses the same signals to rank news versus any other type of content. When Google tweaks those systems and, and tweaks those, those ranking mechanisms, news sites are also impacted, whereas previously news websites pretty much were unaffected by whatever major changes Google rolled out because it was its own separate infrastructure in, in Google's system. So I think that is probably the major shift we have seen. And that also means that it's probably since about 2018, 2019, SEO has become a lot more volatile for news websites. And, and websites have to, have had to work, and news publishers have to work a lot harder to get their rankings and, and earn that trust that Google is looking for from all publishers, not just news publishers. And that for some publishers, they felt a bit of a paradigm shift and maybe a bit of a loss of an entitlement there. You know, when you get a free ride basically into Google search results, like a lot of publishers had been getting for nearly 15 years, and suddenly that changes and you have to work for it like everybody else, it feels like a step back and it feels like an attack. Whereas actually what has happened is Google has leveled the playing field a lot more. You have to earn it now. You have to work for your rankings, no matter who you are, from the New York Times to some, some, uh, news website sitting in, in the, in the damp basement somewhere in the third world country. Everybody has to earn their rankings. Everybody has— is playing again with the same rules. And yeah, that means sometimes you, you know, you, you don't rank for what you feel you should rank for. And that can be a tough pill to swallow, but it means you have to just try harder basically.

Yeah, but don't you agree with me that larger news sites have, uh, advantage over new and smaller news sites? Coverage, uh, amount of production, authority, and so on.

Absolutely. Scale matters. Um, and this is, this is done by design to a certain extent because the larger and more authoritative a news website, the less risk it is for Google to rank that news website. Yeah. There's a reason you see websites like the BBC rank everywhere. It's because A, they have really, really strong brands and, and Google looks at things like engagement signals. So people like engaging with those strong brands and that sends a positive signal to Google that it can show more content from those strong brands. And B, those websites, those publishers tend to invest a lot in proper high-quality journalism. Um, which means they accumulate authority signals, which is what Google also relies on. So yeah, as a big news publisher, you have an advantage over a smaller niche news publisher. Um, and you know, you can use that to your advantage. So yeah, it's not always a fair playing field in that regard because more money tends to mean better rankings, but that's true in every industry, not just the news. Yes. You know, big brands tend to win everywhere.

Yeah.

So true.

Which is why we have, uh, antitrust laws.

So, you know, so, uh, I had this question for later, but I think it makes sense to, uh, to ask it now. Is, I mean, one of the biggest challenges for new, uh, sites, right, is to get into Google News and actually the at large to get it the in the Google News ecosystem, right? And you, myself, and so many other people have been complaining and, uh, expressing their frustration there, right? Because as you mentioned in the past, uh, you would submit a form, there would be a manual review, and someone will decide if you will join Google News or not. Nowadays it's like, you know, um, it's all automated, right? And there's no transparency in the process. Have they ever reviewed my site or not? Or, uh, was I rejected? Are they even aware of me, right? And I think recently, I can't remember when Danny, uh, or someone— I think Danny hinted to there might be a change in the near future around that. I would believe it when I, I believe it when I see it, uh, right? But it has been a big frustration for so many, uh, new and upcoming news sites, uh, especially independent sites that want to join the, uh, Google News ecosystem. Any recommendations?

You know what, this is probably my weakest area when it comes to news SEO because we just don't have enough data yet. Um, you know, when, when Google changed from a manual to an algorithmic inclusion process at the end of 2019, I'm pretty confident no new website got into Google News for at least a year, year and a half, 2 years even. Um, and that's because that's how long it takes to accumulate the authority signals that Google is looking for. We have since seen websites be accepted as publishers, uh, and get traction in Google News, especially in top stories. And on average, I think it takes 2 to 2 and a half years. Sometimes faster, usually 2.5 years or more. Uh, so it's, it is a slow process. Now I think the change that Danny was referring to is that that process may be accelerated a little bit instead of 2.5 years. Maybe it's now a year, year and a half. Um, and Google is, is trying to find the right authority signals a bit quicker, but it is still a slog. It is still something you have to be prepared for. And there is no guarantee either. One of my favorite newly published websites is 404 Media. They started, I think, back in. October, November last year. Um, and they were already getting a really high, high level of brand audience to their website cuz it's a very good technology-focused, uh, uh, in-depth news organization, but no visibility in Google's news ecosystem. Yeah. I think like 5, 6 months down the line now that they should be in there because they have the reputation, they have really good high-quality journalism, but you know, it'll take a long time for them to, to get that traction and start appearing as, as a top stories approved. I mean, the best thing you can do as a publisher is just make sure you focus on building the right authority signals so that Google understands that you are a reputable publisher. And the best way to do that is to publish content that is worth of the, of, of, of the, the label of proper journalism, stuff that people want to link to, stuff that people want to share, stuff that people want to engage with. Uh, and you know, having really good stories around it, be fully transparent about who you are as a publisher. who your writers are, or the, the EEAT signals we keep talking about. And so that when the average user arrives on your website, they realize this is a trusted publisher and they keep coming back to your website.

Yeah.

But the good news is in Discover, you tend to not have to wait that long.

Yeah.

If you write good content and people engage positively with your website, within a couple of months, you tend to get traction in Discover, and that can become a very powerful source of traffic for you. So you're not entirely beholden to Uh, the, the algorithms that define, uh, traffic in Google News and Top Stories, uh, the Discover algorithms have, have a broader remit, a cast, a wider net. So you allows you to get pretty decent traffic numbers if you manage to sort of capture that audience and have that positive engagement with your website.

Yeah, absolutely. And if I may add, I think there are, you should absolutely read Google News guidelines, right? The technical guidelines, the editorial guidelines, and read them thoroughly. Read between the lines. So when they tell you like, hey, we need a clear view who who is the editorial team, right? Editor-in-chief and so on. So when you have like about us and mention your your mission and who are the the people running the site, do you need transparency into that? How people can contact you and all that stuff. Applying like submitting your site to News Publisher Center. I'm not saying any of this stuff is a guarantee, but this is all things that can help you hopefully. when you get automatically reviewed, you check all these boxes. Barry, I want to talk to you about top ranking factors in 2024. You recently wrote an article, Turns Out Google Is All About Links and Clicks. I love this article. We knew this for the longest time. We said it and Google was like, no, we don't take any of these metrics in consideration about links, clicks, and engagement and so on. Uh, and so on. So talk to me, what do you think, right, is the most important SEO ranking factors that news sites should consider and optimize in 2024?

Yeah. Um, yeah, I love, I love that those revelations from Google's antitrust trial pretty much confirmed what we as an industry strongly suspected for many, many years. We always assumed Google was a lot smarter than they actually are. And I, I think that makes sense. You know, I think we sort of tend to overestimate it because it's a big black box and yeah, all this mysticism around it. But it makes sense for Google to rely on signals that its users generate, how people engage with websites, uh, and, and use that in all various different ways to see if that website deserves to be ranked. I still very strongly feel that good links get you to the first page of Google. Good clicks keep you there and get you higher on the first page of Google. And the best way to build links is still to write content worth linking to. So don't do any artificial methods, especially now that Google is again cracking down on the spam industry. And don't use AI to generate content either, because that doesn't attract links either. And then clicks means you have to have a website that is, is pleasant to engage with. And I think when it comes to news publishers that already have traction in top stories and want to maximize that, engagement signals are probably the strongest ranking factors you can focus on. And I almost hate to say this, but yes, that does mean you should not be loading 2,500 ad slots on a single article page.

Yeah.

That does not mean you should be interrupting the flow of an article with all kinds of crap in there that, that will mean people probably just gonna click the back button. It means you have to have a lean, focused website that people enjoy engaging with, where people will, will read as much or more of the content than on any other website, that they have a good depth of reading and that they will click on links after reading the article to read more articles. Around the same or similar related topics. You need to really focus on having an optimal user experience on your website, cuz that signal will be used by Google in aggregate to send more traffic to your website. Whereas if you have a website that people don't like engaging with, that is a negative engagement signal for Google and they're just gonna start dropping your rankings steadily. And then, you know, do you see this, this spiral all the time, John? You've seen this as well, where the traffic goes down on a news website, but they need to keep the revenue up. So they put more ads in there and find more ways to monetize it. And the website becomes even worse to use. And then traffic goes down even more and it becomes a self-fulfilling prophecy where your website becomes so terrible to use that it should be nowhere near the first page of Google cuz it's just an awful website. So sometimes you have to have, I would call it courage in leadership to say that, you know what, we're probably gonna cut revenue in the short term by reducing ad load, but we're going to be looking at building a loyal audience that we can use to, to push the business forward rather than try and squeeze every penny of, of money from every single visitor to the website. If you treat your audience with respect and deliver a user experience that is pleasant for them, they will keep coming back and you will be rewarded with better rankings, more traffic, more visibility, not just in search, but in all online ecosystems where people share news. So you have to have a bit of a long-term vision. and not get too bogged down by whatever quarterly bonus target the commercial team has.

Yeah. I mean, being on the corporate side for 20+ years, right, I truly feel for news publishers, right? Because revenue sources keep diminishing in so many different ways. And it seems the first thing they keep thinking about is ad revenue. Okay, we have less traffic, exactly what you said, let's add more ads to the page. You add more paid, you add more ads, you lose more traffic, and here it goes and so on. And this is the number one thing we have seen over like the last year or two is like more ads on the page, especially the, the ones that interrupt the flow or intrusive ones, will definitely hurt you as a news publisher. Uh, and I think what you said, Barry, about the clicks and engagement even in top stories, I truly believe that the more clicks that you get on top stories, the longer you stay in top stories, right? That's, that's a personal belief. I don't have like a, a, a very good evidence for that, but I have seen it over and over and over.

Yeah, yeah, I agree, I agree.

Okay, so let's talk about technical SEO, right? And is like, what has changed, right? Like if you did an audit 3 years ago and you're doing an audit What are the, the most common or the top 3 things that you look for in an audit, uh, today that you wanna make sure from a technical SEO that these are in a good place?

Yeah, I mean, it's funny because, uh, a couple of months ago I looked back at some of my first audits specifically for news websites, uh, from when I, I started working on my own. And, uh, in terms of the checklist and the stuff I've, I've analyzed, not that much has changed actually. I mean, we, um, I'm still looking for more or less the same things. Um, especially for technical SEO, it's about efficient crawling and efficient indexing. And the rules there haven't really changed that much. Um, because it's still just about, you know, building good websites, really. The things I tend to focus on most for news publishers revolve around those things like optimizing for crawling, make sure that newly published articles can be very efficient— efficiently found and, and crawled. Uh, that you send the right signals about which pages Google should be crawling. You're not creating a massive crawl overhead. And in terms of indexing, it's about making sure that, you know, the pure HTML source, which is what Google will initially rely its indexing on, because that rendering phase where the client-side code is executed is cumbersome and time-consuming, and Google tends not to have time for that in the news ecosystem. So you have to present fairly decent HTML, uh, Uh, very quickly index. Uh, and that's of course, again, about removing obstacles, making sure there's no confusion, having your header tags in order, having your HTML for your article in, in a relatively clean, uh, uninterrupted block. Uh, and making sure that, you know, when you, that, that whole, uh, HTML file gets, gets passed through a parser, nothing really weird happens, but, you know, you get a fairly clean layout of, of where the content sits and that is not spread out over massive blocks of HTML. Spread out throughout the whole, whole webpage. And so that's generally what I focus on for my audits. And, um, we have seen some things shift here and there where new things get introduced. For example, for Discover, having the max-image-preview meta tag, really important to make sure Google can use big banner images for, for, uh, you know, the Discover articles. The same with news article structured data. You know, Google has sort of, I would say, reduced the dependencies there that you can have a fairly lean snippet of news article structured data there. It doesn't have to have all bells and whistles. And in fact, I now found that with smaller, more focused structured data snippets, you tend to get slightly better results cuz Google has less issues of parsing, um, and, and validating those. Uh, and you know, things like author pages, uh, having those links from bylines, uh, having person schema on those that ties into the EEAT where we see increasingly, uh, heavy emphasis on, again, they're not direct ranking signals by any stretch of the imagination. But they are still important secondary signals that we need to consider to make sure that you tick all those boxes and present yourself as a valid, reputable news publisher.

Okay. That's really good. Uh, and I want to ask you, so you, you do all like considerable amount of audits, right? Every year. What are the most common ones that you see? Like common mistakes, common SEO mistakes that you're like, oh yeah.

Um, probably one-third of my audits, uh, I have the same recommendation with regards to load speed and server response time. Not even a third, probably more like two-thirds, um, where the website just doesn't respond fast enough. Uh, sometimes by necessity, if you're a news publisher in Australia and Google crawls from California, yes, you'll have a 1.5-second response time on average, but that's fine because every news publisher in Australia has the same problem.

Yeah.

If you want competitive advantage, use a CDN with a node in California. And Google will crawl you much faster. So, but that's more of a nice to have other than a must have. Um, if you are based in the US and you have a 1.5-second response time on your, on your server infrastructure, you have a problem. You know, Google wants really fast responding websites because it means they can crawl faster. They can crawl more pages in less time. So that's a common one.

200 milliseconds or 600 milliseconds?

They want 200 milliseconds. I think that's unrealistic. I think 500 milliseconds or faster is sort of where I put, put my own limit. If it's under 500 milliseconds, so half a second, If it's more than 500 milliseconds, like, yeah, maybe keep an eye on that. If it's more than 700, 800 milliseconds, like you, you want to improve that if you can.

Okay.

That would be, that's an easy one. Um, I, I am still surprised at how many publishers use tracking parameters on internal links to see how people behave on the website and, and not realizing how insanely stupid that is.

Yeah.

Uh, because what happens is that Yes. Oh, they say we have a canonical tag and the canonical tag will point to the version Google needs to index. But the canonical tag is a hint. It's not a guideline. It's just another canonicalization hint. An internal link to an article is also a canonicalization signal. What happens when people use tracking parameters on internal links to, to track where people click is that, A, you're creating a lot more URLs for Google to crawl.

Yeah.

Especially if these tracking parameters are like very context dependent about where they appear. You can literally have, uh, unique internal links on every single page on your website, which means you're creating so many new links for Googlebot to crawl and they will crawl them because if a URL is one character different for Googlebot, that means, oh, it's a new URL. I have to go and crawl it. There's massive crawl overhead, massive crawl waste. Second, there's a good chance Google will choose to index the URL that contains the tracking parameter.

Yeah.

Because, you know, internal link being a canonicalization signal.

Becoming useful.

And then surface that URL with the tracking parameter when that article is ranked. Which means when someone will click on that link from a Google search, your analytics program will register it as an internal click rather than as a Google search click, which means the whole point of using internal parameters to track clicks.

Yeah.

Is invalid because your data is entirely unreliable. Also, people will copy and paste those links and send them to friends, uh, and share them on social media. Again, you know, traffic from third-party sources will show up as internal links because of those tracking parameters. It is such a stupid way of doing, yeah. There are clever ways to do that. If you want to have tracking parameters on internal links, use hashes. Yeah. Because Google ignores everything after hash and won't index URLs, uh, with, with the hashes. It'll just index the clean part of the URL. So that is a clean way of doing it. Or just use click tracking software sparsely, mind you, because some of 'em can be quite, quite heavy. Yeah. To see where people actually click and how they engage on the website. And most web analytics programs, uh, Will already have a feature built in that, that tells you where people click and how long they engage with your website. So you don't even need to use tracking parameters. And yeah, probably one quarter of the, of the audits that I do will have issues with internal parameters of, of one description or another that create massive call overheads and cause indexing issues for Google as well.

Yeah. So you said, so the common ones, you said the server response, right? The internal, uh, tracking links, right?

Yeah.

Or internal links with tracking parameters. Any, any other common ones that you see?

Yeah. Pagination. Pagination is often done really wrong, either too many pages or not enough pages. Uh, a lot of pagination is implemented with scroll functions, um, or read more buttons that, that use JavaScript. Um, uh, and means Google doesn't see beyond the first page because Google doesn't click or scroll. So any JavaScript-enabled action that requires a click or a scroll is, is not It's just not triggered by Google when they index pages. Um, so we, and we want Google to see more than just the first page of, of articles on any given topic or section page, because we want to show a certain body of work. We want to show that we cover a topic in depth. We have a lot of articles associated with a, a topic. So we need to, to show, you know, decent amounts of, of volume of articles on, on any given topic page. But at the same time, if you have a very old website, with like decades of, of, of archives. We don't necessarily want 8,000 pages of content.

Yeah.

On a topic for Google to crawl and index. Uh, cuz that could be problematic, especially when you combine it with other potential crawl issues, which sometimes are very easy to, uh, to pop up. So you want to find a sort of a happy medium there. I'd say—

What, what's your happy medium? What's your happy medium? What's your number?

I'd say at least 10 pages of content with at least 20 articles per page. So we're talking 200 indexable pages. Associated with a topic page minimum. Uh, uh, more great, but you also wanna cap it. I say if you want to absolutely have a cap on it, say, say 100 pages. Yeah. 100 pages of content maximum. I'm okay with, with capping it at 10 pages as well. You see some, some publishers that I work with, like, uh, Future Publishing Tech Radar, they have 10 pages and then they go to a database archive page, which I love. Yeah. Very elegant solution. Yeah. Others like your former employer, Condé Nast, on Wired, they have like a next page. Uh, so there's only one pagination page to the next page at the bottom of any given page. And because of how Google prioritizes crawling, you lose link value with every hop. So every deeper pagination page becomes less important for Google and, and less of a priority for crawling. Uh, and there's no limit to that in terms of, you know, just those next pages go on for as long as they're pages. But I'm not worried about crawl waste on that because of the way Google prioritizes And I find that a very elegant solution as well. Yeah. So I say if you have like pagination links at the bottom, you can do either one on every page, like next page, previous page, and just leave it at that. And that can run for as long as you have pages. Or if you have like page 1, page 2, page 3 clicks, et cetera, I'd say you want to put a cap in there between 10 and 100, depending on the volume of, of content that you publish and the authority that you have. Um, and there's no need for Google to look beyond 100 pages of content. That's just a waste of everybody's time, cuz no user is going to do that either. You know, users will scroll down a few pages, but no more than 4 or 5, and then they're just gonna use the search function, aren't they?

Yeah, no, I totally agree. Okay. Let's, let's talk about speed a little bit, right? There are so many terms thrown out there, right? And a lot of SEOs sometimes get confused. Speed, server response score, web vitals, and so on. So, uh, what is speed? Among all these terms, if we're going to talk about server response, Core Web Vitals, right, and whatever other metrics you want to discuss, what do you think is the most important? Because Core Web Vitals were sold to me, to so many SEOs over the last— like, it was business in itself, right? You, you go to a site, it's like, you know, your Core Web Vitals are so bad, you're losing so much. And it became a business over the last 2, 3 years, and people were making so much money. So I would love to hear from the master. Core Web Vitals, speed, uh, server response.

What's your ultimate goal? The one I care about the most, and it's not even close, is server response time. That has a very direct, very measurable impact on crawl rate. If you improve server response time, you improve crawl rate, you improve how many pages Google can crawl in, in any given point in time. And that's the one that I think has the biggest impact on performance of a website. terms of crawling, indexing, and, and ranking on Google.

Yeah.

Core Web Vitals is an interesting one because, and I'm maybe a bit of a conspiracy theorist here, I'm not actually sure it's a direct ranking factor. And I, I put emphasis on the word direct there. I think I am a fan of Core Web Vitals because I think it is a very measurable way of, of measuring, um, user experience. It gives us metrics that give an indication of, of how pleasant a page is to interact with. There's some things you cannot capture in Core Web Vitals, but a lot of things you can capture with Core Web Vitals. And I think it's a very clear way to give, uh, designers and developers measurable targets when it comes to optimizing the design and interaction of webpages on your website. And so in that regard, I am a fan of Core Web Vitals. I am not convinced they are a major ranking factor or even a small ranking factor because we see websites with Absolutely terrible Core Web Vitals perform quite well in all Google's search elements. Um, but I do think they indirectly matter because like I said before, bad user experience, which generally also means bad Core Web Vitals, means less engagement with your website. And I think in that regard, a lower engagement means less, uh, positive signals for Google to work with. And that means you can, you know, lose rankings over time. So I do think it's something that publishers need to take seriously. I don't think it's a major, uh, ranking factor in and of itself to have bad Core Web Vitals.

Yeah, I totally agree with you. And I, like, I'm— I spoke with so many publishers where, like, they're so sold on the idea, it's the most important thing that they have to do for their roadmap, right? And the content is bad, the architecture is horrible, but Core Web Vitals is like the number one, and they do all this work and then they see no impact. I think Core Web Vitals could be a tiebreaker, like if the whole industry or the whole country or whatever has a really good Core Web Vitals and you are the only one who are like left behind, enhancing your Core Web Vitals definitely may help you. I saw Core Web Vitals do some impact in Discover, uh, a little bit, right? So, but I totally agree with you. I think it was, it was overrated and oversold to so many sites.

And even when it comes to Discover, I'm not sure if it's Core Web Vitals or just engagement signals, because bad Core Web Vitals did lower engagement and engagement is a very high factor in discovery.

Okay.

More so even than in regular search. So yeah, that's why I'm a little bit like, okay, Core Web Vitals, I'm a fan, but I'm not sure it's actually a ranking factor. But I do tend to mention it in most audits because I do feel that publishers need to take it seriously because like I said, it's a measurable way of, of looking at your user experience.

Okay. Thank you. Uh, guys, a reminder, if you click in Q&A, please add your questions to Barry. I see like 2 or 3. good questions there. You can vote on these questions and we will, uh, answer all of them. I hope so. Okay, Barry, uh, let's move to another thing. Um, large site audits and large site migrations, right? When, when you see a site that is like, you know, 2 million, 3 million, 4 million pages, right? Uh, what would you advise SEOs that doing these kind of audits in-house, where to start Like, what's your process?

Uh, it's always based on templates. You might have 10 million pages. You probably only have 12 at most templates. Yeah. As in homepage, section pages, topic pages, articles, and maybe 3 or 4 different article types. Those are your templates and you optimize on a per-template basis initially. That's, that's stage 1. You look at all these templates and you see, right, what can we improve for this type of page? Um, in terms of rapid crawling, rapid indexing. Um, second, you don't necessarily need to crawl an entire 10 million URL website. Uh, you can take a sample there. I would say it needs to be a significant sample. I tend to crawl 1 million URLs on every audit that I do to a max. Um, and, but I find that that's enough. That gives me enough data to work with to sort of draw conclusions about what would apply to the entirety of the website. Um, it also helps if you divide the website up into chunks, like, okay, I'm doing sports section this time. Or I'm just doing a local news section in that time and limit your call to that. So especially if you have like interesting technology stacks, one of my older publishing clients has like 3 different technology stacks running the same website because they power different parts, literally different sections of the website are powered by different technology stacks. Now those are interesting challenges, I can tell you.

Yeah.

It's quite rare. Usually a single hostname runs on a single CMS. Uh, and then it's looking for patterns, uh, patterns in internal linking, patterns in, in, like I said, uh, article templates and things like that. Um, and also look at what Search Console tells you. You know, Search Console is actually quite a good source of information, not necessarily to point you to fixes, but to send you on the right path.

Yeah.

I think Google Search Console is very good at pointing you in a certain direction, but it doesn't tell you what the actual problem is. But if you have a typical Search Console pages report and you see certain things like pages with redirects or soft 404s, or, you know, uh, uh, uh, indexed or chosen a different canonical than the canonical tag, and that points to an issue and you then need to explore what the actual root cause of that issue is. So, uh, with those 3 sources, looking at page templates, Uh, a limited 1 million URL crawl of a website. It can even be 100,000 URL crawl if you have, uh, trouble with, with crawl budget, uh, on, on your crawl tool and looking at Search Console Pages report. That sort of is the starting point for every major audit, no matter how big the website actually is.

Okay. That's very helpful. Uh, if you don't mind me asking, what's your favorite, uh, technical SEO tools that you use? I mean, there is great tools out there, but what are the top 3 that you go to, uh, with most of your audits?

Um, I'm, I'm trying not to rely too heavily on tools, but the ones I use every single time, Sitebulb is my crawler, especially now they have a cloud-based version, so you don't have to worry about, uh, hosting it, uh, yourself. And because they have very detailed reports on, on, on technical analysis, and I just love that about it. Google Search Console is my go-to for pretty much any, any starting point from audit. And I consider it a tool because of the data that, that it has. Yeah. And beyond that, you know, I use things like WebPageTest for a fairly detailed load speed analysis on websites. And I just use tools as and when required. So I'm not— I try not to be too heavily reliant on tools. I use as little as I can, because I'm still one of those old school guys that I don't want to be caught looking at a website purely through the lens of a tool without actually looking at the website. I think we often get lost in the data that tools present to us, but I think we just need to actively look at a website and engage with the website and understand it before we can draw conclusions on how to improve it.

Yeah. I think the last point you mentioned is like, it's so crucial. I have seen so many SEOs that doing audits, uh, either in-house or, or like, uh, as a contractor that they don't even look at the site. They don't browse the site. They don't look at the articles and they trying to understand understand the site completely from a tools perspective, which I think it's, it's a, it's a big mistake. Okay, so we're almost there. I want to talk to something dear and near to my heart— syndication, right? And a year and a half ago, I put this big article out there in NewsDash, and I said, um, Yahoo is eating a lot of publishers' lunch, right?

And I had a lot of clients talk to me about that article. That prompted a lot of conversations.

It was, it was bothering me for a long time, right? And I wasn't saying Yahoo is the problem here. I was saying, hey, Google has been advising us for the longest time to do certain practices, you know, canonicals, link to the original article, all of that. And many of the publishers have been doing it, but it's not working. And Danny came and we had like a conversation on Twitter back and forth. A lot of people interacted with that conversation and, uh, it And then Google said, here is the ultimate answer: if you don't block the article, we cannot guarantee the canonical. Canonical is just a hint, especially when it comes to syndication. And it was very clear recently, over a month or ago, is like maybe one of the reasons is Google doesn't want to reward the original site if the syndicating site is getting all the links and the traction. So why they should transfer all of this So what's your POV on the matter?

Yeah. Yeah. Always a fun topic. I mean, it, it points at fundamental, uh, weaknesses that Google has. Uh, again, we tend to overestimate how smart Google actually is. Um, Google sucks at deduplication. Deduplication is hard. Um, and we have to keep in mind when Google sees 2 webpages on different websites with identical content, for us as readers, we can fairly easily determine, oh, these are the same pages. Google doesn't necessarily have that same capability because if it looks at the HTML source for both, uh, webpages, they'll be very, very different. Different design, different navigation, different boilerplate content.

Yeah.

Uh, so Google doesn't have a lot to go on and Google then needs to be able to extract pure on-page content from both of those pages and then compare those. Now that requires for Google to render both of those pages in a browser, uh, extract the full DOM and then see, you know, oh right, this is probably headline here, headline there, and let's see how on-page content is different. That's very time-consuming. That's very hard for Google to do. And in news, Google doesn't have time, therefore very limited deduplication. In fact, I'm pretty confident the only deduplication Google does in the news ecosystem is headline deduplication, where they don't have articles with identical headlines show up next to each other in top stories. Uh, uh, but the rest of—

I saw many examples, Barry. I saw many examples where 5 headlines are exactly the same, and they're all the same.

So they don't be, they're not even good at that. Yeah.

No, but you're right. But it's, it's to a big extent that, but I have seen examples here and there. Yeah. Sorry to cut you off.

No, that's exactly, that's it. They just suck at deduplication. So Google wants you to solve that problem for you by just no indexing syndicated articles. Um, now I have to give credit to Yahoo. We had a really good conversation with actually Yahoo employee at the, the conference I was at last week. And they said, yeah, we don't necessarily want to steal your traffic. Um, you know, if you send us your traffic, then by all means put a delay in there, you know, give it to us late because we don't really need that traffic. We want it because we want to present like a news homepage for users and they can read all that they want from the Yahoo News homepage. But Yahoo doesn't necessarily want to rank in top stories because that's not going to be the primary driver of traffic. They rely heavily on brand awareness. Uh, traffic and people using Yahoo as their homepage and then using that as a portal to read all the news. So people never get tempted to go to Google search to find the news. They can just stay on Yahoo. That's why Yahoo wants, wants your syndicated content so they have that full experience for users there. But Yahoo, by virtue of being a highly authoritative publisher, will steal your top stories traffic the moment you give them your articles. Uh, you know, that's, that's just how this system works. Um, so noindexing is a way to get around that. Now Yahoo is probably not gonna say yes to noindexing unless you are very powerful publisher and they desperately want your content. But for most publishers, the best you can get is a canonical tag back to your own website. Uh, and Google will, will ignore the canonical because the difference in authority levels will be way too massive. And Yahoo is just gonna rank the moment they publish that article. So build in a delay to your feed, even 4 hours, even 8 hours if you want. Actually, 4 hours is probably a good shot. Publish your article on your own website and don't send it out on the feed to Yahoo until 4 hours later. Gives you a good chunk, good window of opportunity. To get, uh, traffic in top stories, uh, pointing directly to your, to your own article. What you could also do, and again, Yahoo knows this and sometimes even advises clients to do this, is to send unoptimized versions on the feed. So headlines that are just not as well optimized, that maybe even lack the right keywords, but maybe a bit more clickbaity headlines, uh, that you send to Yahoo, whereas your own headlines on your own website, they're maximized for, for search performance, very keyword rich, very descriptive, uh, and, and, uh, sending all the right signals to Google about what the topic of, of that article is. You can do that too. And another scenario which I've seen is where you send a slightly less complete version of the article to Yahoo. Yeah. Yahoo gets the headline and the first 2 paragraphs, and you get the full, full article on your own headline, subtitle, and 7, 8 paragraphs of text. Again, if, if you are open with Yahoo about this, then they'd probably be, be okay with it, you know, cuz again, they don't necessarily want to steal your search traffic. They just want that content to fulfill needs for their users so their users never get tempted to go to Google to search for the news. So, you know, there are different ways you can optimize the, the, the syndication challenge, uh, and, and give yourself the best opportunity to rank in, in Google without, you know, your traffic getting stolen from under you.

Yeah. Um, I agree with all the points you said, and again, I'm not pointing fingers at Yahoo at all, right? Yahoo is just syndicating the content. But I have to say that a part of the revenue that publishers get, it depends on how fast you— they syndicate their content to their partners, right? So when you syndicate your content immediately, it's a different revenue if you syndicate it after 1 hour, 4 hours, and so on. So while it's completely valid advice, but in practicality and like business practices and contracts, it's a different revenue, uh, uh, Yeah.

And you sort of have to then make a commercial decision, you know, the potential search traffic versus the Yahoo revenue. Yeah, exactly.

It's at the end of— it's a trade-off. Okay, so we came to 45 minutes. Uh, we're gonna take a quick presentation from our sponsor NewsDash, uh, for 2 minutes, and then I have really exciting news for you guys. Uh, I have a special discount on the upcoming NEST where Barry and myself are putting the agenda right now, selecting the speakers and the topics and so on. This is our 4th edition, and then we're going to continue with questions, live Q&A with Barry. Let me invite to the stage Dave. One second here. Show on stage. And I'm gonna hide from stage me and Barry for a second.

Thanks, John. All right, I am now going to show a brief overview of NewsDash.

Okay. Cool.

All right, here we go. So for all those who don't know what NewsDash is, we are a real-time news SEO software. Uh, we scrape every 15 minutes, so compared to conventional evergreen, uh, tools, we are truly a real-time SEO platform for publishers. Some of the things we help with is what should we write about, how big is an opportunity, ranking, own performance versus competition, optimizing for keywords and events, and getting a good idea what's happening in the landscape in real time. Once again, let me update this information every 15 minutes. These are some of the clients that we currently work with. All right, so starting off, one of the biggest things that we think is important as NewsDash is understanding the trends that people are looking for. In Google Trends, you have the score from like 0 to 100. In NewsDash, we can actually give you the estimated search volume to kind of show a replication of what that would be on ROI from writing that. And as well too, as we're showing you top new trends, the trend overview, and any context— why is it trending, kind of to the point of what they're saying, looking at the article as well too, just not just a tool. Brand Visibility. This is showing us once again how our competition is doing. This is giving us an overview of every hour. For instance, we see these peaks, you can click on any of this to see exactly the articles that your competitors are ranking on. We can do 2 years of historic data. We can show you all different SERP features. So if you want to do live blogs, live red pills, we can do unlimited competitors. And most importantly, We have automatic clustering, so if we're looking for elections, the keyword Joe Biden would probably pop up as well too, because we're able to deduce what keywords are relevant with our keywords. From here as well too, we can also show you a general overview of your news topic authority. This is giving us a very nice insight of our keyword shares, exactly showing us what our competitors— why they're the number 1 spot for the last 30 days. It shows us the amount of ranked URLs they have and the average position they are when they're in top spot. Now, because we're— this is for OnBrand, this is a technical SEO webinar, we also have a feature showing us for indexing the published articles. So you can see here the blue is— and the blue and the red, this is pretty much in line. All of the articles are pretty much being indexed and published, so not a problem, but just gives you a good idea, some technical insight into how what's happening for the last 30 days. And then ones that are not, we can get a good idea of published URLs that are not indexed. So any technical issues, NewsDash can help you pick that up and give you a good idea of what's not being indexed and putting that into a report for you. Finally, as well too, we also have another tool called GDash where we're able to do some Google Discover feeds tracking as well too, just for the cherry on top. And that's everything for me. If you guys have any questions, this is my email address. Feel free to reach out, and I hope you enjoyed the webinar.

Thanks a lot, Dave. Uh, this was very helpful. Let me bring Barry again to the stage. Uh, I'm very excited about the last feature. This is something we added to NewsDash where you can actually track how long it takes Google to index your content after you publish it. And the way we get it is like we track your XML sitemaps for Google News and see if these articles are in Google or not. So it was very like fun feature to build. Another thing I want to say is this is the article I wrote about the news syndication, and I listed 10 different things that you can do. It's interesting, Barry, you mentioned a couple of things that were controversial to some people that I mentioned in the article as well, like sending less optimized version of your content to Yahoo and so on, how to measure the trade-off and stuff like that. So it's, it's a fun, uh, fun read. Okay, Barry, we have a lot of questions, so let's get, uh, to them. Uh, our first question here is, uh, this is from Simon, uh, SEO director, Future PLC. Hey Simon, uh, on large sites which have a lot of old news articles that are no longer relevant receive traffic, should this be, uh, tied up and redirected, or is it better to leave it there because of topical authority, external links, and so on? And let me add to this, there was a lot of discussion last year about pruning content from the site, right? Hey, we gotta remove, uh, 60% of all the content that we have on the site. And then Google said, ah, wait, wait, it's not such a great idea.

So I'd love to hear your Yeah, there were a few case studies I think published that, oh, we deleted 60%, we saw a traffic increase and people were jumping on that bandwagon. Um, I'm pretty sure that those case studies don't actually have a direct correlation between or causation between removing articles and, and increased traffic. Um, but that, that's a side point. Um, I have fairly strong opinions on this. I think you should not delete content unless it's entirely, uh, outdated. Um, I think that once you publish an article on your website, you are basically telling Google, hey, this is content that we wrote, that we, uh, uh, have investigated. This is like journalistic output. And even if the article then disappears into the archives and, uh, stops being relevant, I still think it serves a purpose. There's a purpose because it's a point in time and a, a news report in time. So it's a historical document to a certain extent. And it's also a point in time. And also because it, it, it adds to your, your topic authority. You're telling Google, hey, this is part of a whole history of content around this particular topic. I would say the only reason why you should want to delete old content is because you maybe evolved your editorial output a little bit and you're sort of ashamed of that content as in it's, it's not good content, it's bad content. It's, it's not necessarily something you'd write today. Uh, and it doesn't have any, uh, links pointing to it. It doesn't have any. In which case I'd say, yeah, okay, maybe you can delete that. Or, um, I read an interesting one the other day where a news publisher was considering deleting a lot of old content because it just, it, it, it was too expensive. The, the, the sheer volume of articles, it cost too much money to host. In which case they had to make a decision, right? How, how long are we going to keep stuff? In which case you probably want at least a decade of articles before you start pruning them. Uh, and, and, and, uh, And that's, that's considered probably long enough, I would say, to, to maintain topic authority. But in almost all cases, the problem of old content is an imagined problem. It's not a real problem. I don't necessarily care how many articles you have on your website. I do care about the, like I said, crawl efficiency, pagination issues, uh, that tends to be associated with old, old content, but is actually a separate issue and much, much easier to solve. Um, I don't think you should go through the whole exercise of pruning content just for the sake of pruning content. It really doesn't have any benefit to do that. You should only go through that exercise if there's a different goal you want to achieve by making your website faster, by saving hosting costs, by improving crawl efficiency, or by, you know, removing content that doesn't align with your topic authority. Um, but the, the sheer point of having old content isn't a problem in and of itself. And therefore isn't a, a problem that requires a solution.

Okay. So the, the top reasons I heard is like, hey, John, it's mainly because we wanna improve crawl efficiency, right? It was like, we have millions of pages and that's the only reason.

Um, and, and they don't even know how often those pages get crawled. Oh, crawl efficiency. No, no. Google is actually quite good at prioritizing the right URLs to crawl. And yes, Google will occasionally crawl old URLs. That's the nature of the beast. They haven't drastically shifted around. Those URLs still exist, but Google will focus its crawling on your most important pages. And an article published 15 years ago is not one of your more important pages. You don't necessarily have to worry about Google wasting crawl effort on that. They're going to crawl your homepage, your main section pages, and your newly published articles. That's what Google is going to crawl. Do you know you have a crawl issue? Are you having delays with indexing of newly published articles? Are there massive errors in your content? If you don't know any of those, you probably don't have a crawl issue.

Yeah. It's funny because the first question you would ask, do you have access to the server logs? Do you know what Google is crawling? I was like, no, not really. But we looked at Google Search Console in the, uh, report and it says it's only 25% discovery and 75% refresh. If I, if I remember the—

25% discovery is quite high, to be entirely honest. You know, if you're more than 10% discovery, then you're probably, you might have a crawl issue in a different way. But it's just, it's, it's such an idiotic imagined problem. You know, we have much more important actual issues to solve than imagined problems on top of that.

Yeah, I totally agree. And before I forget, guys, this is the special code, uh, from Barry and I. I'm gonna show it here on the, on the site.

Show.

Okay.

Okay. I don't know. Can you guys see it? Special code for Ness, BA, Barry Adams, Webinar 25. So you get 25% on the upcoming Ness of 2024 if you go to newsseo.io. Newsseo.io.

Just put it in the chat, John.

Oh, thank you. So let me show this one too. Uh, okay. Uh, and, uh, it's valid for today and tomorrow. So this is only special for Barry's, uh, webinar. Okay.

Let's go.

You deserve more discounts, Barry.

Okay.

Let me, uh, bring the next question. Um, Barry, AI, uh, right? What are the best AI techniques that you have used for technical SEO performance? I love that it's for technical SEO. Good. It's not— we're not talking about content here. And it feels like most of the tools that we're still using— and I'm a big— I am a big fan of Sitebulb, uh, still have not yet utilized AI. Uh, maybe Botify did some of that stuff. I'm not sure yet, but I haven't seen a lot of tools start like, you know, utilizing AI in the reports and, and helping SEOs like quick pattern, uh, detection and stuff like this.

Yeah.

What do you think?

Yeah, I think that would be the, the, the easiest scenario is AI and machine learning systems are good at pattern detection, so they can help you with, with analyzing large websites. Um, I'll give you a real life example that I saw not too long ago was site mergers where, um, 5 sites got merged into one and these were all, uh, formerly competing companies that were acquired and merged into one company. And so they have a lot of overlapping content targeting similar and the same keywords. Um, and doing manual merging was probably not feasible for the scale of these websites. Uh, but they did want to preserve the content and wanted to preserve the, the ranking value they had associated with it. So they used AI systems to basically analyze content on a page level, find, uh, semantic equivalence, uh, and make those mergers semi-automatic with a little bit of editorial oversight and automatically implement redirects. Uh, so I thought that was a clever use of AI that saves an awful lot of man-hours having to do manual redirect management. Mapping or even semi-automated redirect mapping. And also automating content mergers to make sure that it sort of ticked all the boxes and the semantic value for the merged page was maxed by the previous 5 pages that it replaced. So that would be a smart use of AI to help with site migrations and replatforming and site mergers and things like that. Having said that, I'm old school enough to always be a little bit wary of new technologies. And again, we've I don't like looking at websites purely through the lens of a tool, be that an AI system or any other tool. Um, so I would always prefer a manual-first approach to solving technical challenges. Well, the AI system might say one thing, but that might not be the actual problem. That might be a symptom of another problem. And you won't know that unless you start investigating that problem yourself and doing the hard graft to look at that. So I'd say always be a little bit wary of AI tools that promise to solve all your technical SEO challenges. They're not going to be able to, but they might help you along the way and save you a lot of man hours along the way.

Okay, absolutely. Um, I was hesitant to ask you this question. I had it as one of my questions I had on, on my list, but I'm glad that Matthias asked this question. So this is the question. I see a lot of pages artificially refreshing their news articles. By generating new URLs and changing publish date without changing the content much or at all, is it a dangerous practice? Let me add to this, Barry. I have seen this practice a lot in Europe. I don't see it much in US and other places, but in Europe, and I'm not going to mention specific publishers or countries, but it's like the same URL will be refreshed maybe 10 times during a single day, right? Its ID Change slug, change slug, change slug. And it's about a year ago I was speaking with one of the top Google engineers and he said, oh, we are aware of that for, for many years and we hate it and we don't like it. We haven't just taken an action against it yet. And I would assume that because everyone is doing it in Europe, maybe so. What do you think? What's your opinion here?

I think some of those clients, those publishers might be clients of mine because it's a known weakness. It's a weakness in Google. URL means new article in Google's news ecosystem. Like I said, they don't have the time, uh, to do deduplication properly. So they're just going to, you know, publish an article when it's on a new URL as if it was a new article. I genuinely think that's fine if you do it in moderation. I have seen an example not too long ago. I think it was the middle of last year where a website did this way too much. Like you said, 10 times a day, and they did it for like every single article that they had.

Yeah.

And, and, uh, let's just say that an algorithm update in the tail end of last year was not kind to them. And I think this may have played a role in there. I think as a tactic to change the URL and maybe tweak the headline a little bit to give an article a boost in top stories and get it to be reinterpreted as a new article is fine if you do it for articles that you absolutely desperately want to keep in top stories. So you feel like this is the best piece on, on this. We need to keep this top stories piece. And let's be honest, if you, if that's once a day, you're already doing really, really well. This is a tactic you should use sparingly. I do think it will eventually raise flags. What that guy said to you, uh, or, or person, whoever they were at, at Google, is, is we, we're not taking action on this yet. I think the keyword there is yet. And I think I know at least one case where Uh, cuz it is an exploit. It's a weakness. Google doesn't like it when you, you poke at their weaknesses. Uh, and, and that stuff tends to have repercussions. So use it, but use it sparingly. Don't go overboard. And, uh, and if you see others used it aggressively, don't feel you need to chase that. All you need to do is stick to your own guns cuz there will be an algorithm update where those websites are gonna get slapped in the face so hard that they're not gonna be able to do it anymore. would know which way is up or down anymore. So you're better off sticking with what you know is the best practice than trying to chase after the cheap clicks just by using those tactics.

Yeah, I think the best analogy that I heard, I think it came from Lily, is like, it doesn't mean that you can get away with some loopholes and taxes for so many years that it's always the right thing to do. Eventually, It will get you.

So it works until it doesn't, and the pain that follows the it doesn't tends to outweigh the benefit you get from it in the it works phase.

Yeah, and I think Cindy also did analysis on short-term gains versus long-term. So yeah, I think if you do it for what matters in a conservative way, I think it's fine. Absolutely. Okay, we have another question here from Katrin. What do you think is going to happen with news with SGE? What's your POV here, uh, Barry?

Oh dude, if I had an answer to that, I'm not entirely sure. Um, I don't think SGE is gonna drastically change search though. I think people are going to use it in specific contexts. I don't think news is going to be one of those contexts, not a dominant one anyway. I think people still want to read the news from their preferred news sources and will keep using the established channels. I do think SGE is very useful in summarizing news up to a certain point, and I think Google desperately wants to keep publishers onside to improve and train their LLMs to stay up to date on current affairs and increase the accuracy of the output of the large language models. So there's a very good incentive for AI suppliers like Google and OpenAI to keep publishers onside. To keep publishers onside. And at the moment, the exact opposite is happening. Publishers are actively blocking AI systems from crawling their website and being trained on, on their news output. So I see it as an opportunity for publishers, especially the larger publishers, to create commercially interesting engagements with large language models. For smaller publishers, I don't think the SGEs are necessarily going to steal your content. However, if you have a publishing strategy very heavily reliant on evergreen, um, like top 10 lists, recommendation lists, that sort of stuff, you might be in trouble. I think SGE could steal a significant chunk of your lunch there. Um, because, you know, people will start using it, I think, for that sort of content. But on the other hand, I'm hearing reports from people who are in the SGE beta that they sort of almost stopped using it a little bit because it's just not really adding much value to their online experience. So they sort of play with it. I think maybe 10% of people in the beta actually use SGE with any sort of regularity. So maybe it's one of those cases where, like with all the new technologies, it's massively overhyped at the beginning. We're now seeing the downturn. I think it will pick up again. I think there will be a very defined, clear use case for large language models. I don't think it's going to drastically rewrite the entire landscape of the web. And that is an evolution of my perspective. I'll be entirely honest with you, Because when LLM first appeared, I thought, oh shit, this could really be a total game changer. But now I'm like, no, it's actually more of an evolutionary step. And yeah, some things are going to change, but a lot of things are going to stay the same.

Yeah, I think it's even evolving for Google itself. I don't think Google made up their mind yet which direction, right? It started like huge and then start shrinking and shrinking the experience. It used to be on so many queries and now less and less. Uh, so I'm not as concerned as I used to be, uh, like you, like last year. I, I think it's not the all-in-one kind of killer for all kinds of traffic. Um, I agree with you. Okay. Um, another question here from Antonina. Uh, what is your advice on categories for a niche new site? Are you— are broad categories better than specific ones? Should we use subcategories to break down topics and to what extent?

Yeah. Oh, that's a good one. Um, yeah, you can get really, really niche sometimes. Um, I am a fan of actually having main category pages be fairly specific, not just say news as, as a category. That's, that's bloody useless. But to be, especially if it's a very niche publisher, to maybe use specific terms, specific subtopics in your top navigation. That are directly applicable to that industry. And that makes sense within the niche that you're focusing on, because those already send very strong topic authority signals to Google that, oh, these are the topics that you covered. So you are a specialist on this broader topic where these subtopics are part of the overall entity of that main topic. I think it's more of a matter of how deep do you go? Uh, and that, you know, relates to also creating tags, for example. Because let's be honest, a tag page is just a category page that doesn't fit in your top navigation. That's how you should look at a tag page. It's a way to collect articles around a specific subtopic that you write about. Um, and when it comes to tags, I think if you have fewer than 5 new articles associated with a tag in any given year, it probably is not worth having a tag. Uh, because that's just maybe not enough, but it also depends. I think maybe too easy of an answer. That also depends on the volume of content that you have. If you produce, you know, 20 articles a week, then maybe 5 on a tag is enough. It's more than enough. And if there's value in having that tag, then by all means keep that. So it's a bit of a shortcut answer there, but it's always relative to the context in which you look at that website. But yeah, I don't— I'm not a fan of using too generic top navigation labels. I think you can be relatively specific without being overly specific. So that's, it's also, it also serves a purpose to your audience. When a person arrives on your website that, that, you know, is your target audience and they understand the industry that you write about and they see using the correct terminology in your top navigation, they know they've landed on the right website. This is a specialist publisher on, on that topic. Whereas if it's just like news and reviews and latest, yeah, that's useless. You know, be a bit more specific, be a bit more focused on what you want to promote. So yeah, I wouldn't recommend falling back on just general labels on your top navigation.

Thank you. Uh, another question here. Uh, please talk to us about the SEO metrics that measure effectiveness of the editorial team. And this is from Andriy, probably I butchered your name.

My apologies. Um, by the way, first, Andriy, you're doing great work. Love your publication. Uh, editorial effectiveness. Um, I mean, the bottom line's always in the clicks, naturally. Um, I am hesitant to, uh, measure the performance of an editorial team based on the performance of just one channel. I don't think you should measure a journalist's effectiveness based on how many clicks they get from Google.

Yeah.

Because that will lead to a race to the bottom. I think you need to have a holistic view of, of the editorial room in, in general and use different metrics across different, uh, uh, uh, channels to decide whether or not, uh, content is doing what it's supposed to be doing. SEO content should be part of that, obviously. And the way you measure SEO performance is very straightforward. It's clicks from Google. It's just clicks from Google compared to a baseline on your website. You should, you should be able to measure that. Fairly confident about what an average article on your website should be getting. Again, you need to place that in context. If it's a niche topic, it's gonna get fewer clicks. If it's big general news, it's gonna get more clicks. Um, but you should have a baseline maybe on a per-section basis, even like sport articles on average getting this many clicks. Uh, local news gets this many clicks, politics news gets that many clicks, and that's sort of a baseline. And you can aggregate that across all Google sources. You can split it out on Discover versus search. I would not recommend splitting it out even further than that. Yes. You get lost in too many details and it doesn't really matter to the big picture. It's purely about clicks from Google and it's irrelevant if it comes from a top stories click, from an evergreen click, from a news tab click or a Google News click. A click is a click. Uh, and most of it will come from top stories, hopefully. And then Discover as a separate one, I would recommend splitting that out because optimizing for Discover is slightly different than optimizing for search. So I think you should treat them separately. But in the end, it's just how many clicks you get. Um, and you know, that can be very easily related to using the right phrases in the right places in headlines and subtitles. Uh, and you know, um, other things play into it in building more long-term value and they are harder to measure, like good use of internal links pointing to topic pages, things like that. They don't have a direct instant correlation with, uh, clicks from Google search, but they do help you build more traffic over a longer time. And that's not something you necessarily can measure on a direct one-to-one basis, but it needs to be good habits. And, you know, editors and journalists need to feel empowered to do that and encouraged to do that. And maybe sometimes you do like a refresher, like once every few months where you look at maybe getting a news desk on a specific topic together and say, hey guys, let's make sure we keep these internal links in. Let's make sure we stick to these best practices just to make sure that it stays front of mind. of center, front of mind for the content creators. But the only metric really that matters is clicks, really. Yeah.

And I think what you said about what type of content it is, right? Sports will get different average clicks from Google versus entertainment versus opinion versus futuristic pieces, right? So it's all different. Okay, let's have another question here. This question is from Harsh Better Collective. If a publisher is approved on Google Publisher Center, they do appear in Google News tab, but never on Top Stories. What can be the technical issues, uh, that may stop the articles from appearing in Top Stories? Any checklist that one should follow?

Uh, probably not a technical issue. Um, probably it could be any one of a few things, but it's most probably not a technical issue. It's most likely that the publication is too new. The News tab has a broader remit than Top Stories, uh, and, and websites will appear in the News tab faster than they would in Top Stories. And also the News tab is more than just pure news, uh, publishers. Google casts a bit of a wider net in the News tab and allows websites with slightly different backgrounds to, to perform there as well. But it's a fairly small traffic source, so it's not exactly going to, uh, shift the needle for you in a, into a dramatic way. Um, or it could be that your website just doesn't have the right authority signal, that you lose out to websites that are more authoritative than you in top stories boxes, in which case get better. Uh, which is very casual for me to say, but very hard to do naturally. But yeah, you need to invest in proper high-quality journalism to build those authority and engagement signals to become one of those 5 to 10 websites that you see appearing in top stories all the time. Um, but it's probably not a technical issue. 'Cause if you're already showing up in the news tab, Google will already, uh, index you. Um, having said that, there might actually be a technical issue in that there might be a delay in indexing, uh, 'cause top stories has a, uh, a preference for newer articles versus older ones. Um, and if Google doesn't index your article until it's already a few hours old, it might be fine for the news tab, but it might not be fine for top stories, in which case you need to look at how fast Google crawls and indexes your website. And if there's any obstacles that are, uh, are involved with that, like for example, maybe your website uses client-side JavaScript, in which case, yes, you will show up in the news tab, but no, you will not show up in top stories.

Uh, Barry, can I add also that I think, uh, let me know what you think. I think being accepted in the News Publisher Center doesn't mean you're accepted in Google.

Oh no, it's entirely irrelevant to your visibility in top stories, but he said news tab. So I think he already has some presence in the Google News ecosystem, just not in top stories.

Yeah. Yeah. I think the question is that, hey, we're already in the News Publisher Center. Doesn't mean a thing. Right. So, and yeah, I shared an article because a lot of people use site: to check if they're in the news tab or in Google News and stuff like this. So there's a couple of easy steps. The easiest one is to go to Google Search Console and see if you get any Google News traffic to begin with or not, right? But there are some easy steps to, to confirm that you have a publisher page in Google News. And then the News tab, I recently tweeted too that less than 2% actually, uh, visit that tab at all. It's Top Stories and Google News, I think, the most important, and everything that Barry just mentioned right now. Um, okay, I have to ask you this question because it's your favorite topic, but I think it's Uh, okay. What is the future of AMP for News?

Uh, I'll give you a one-word answer. None. I'll give a bit more background. Um, we all know AMP and we all hate AMP. Look at AMP as open source. They have a GitHub repository. And all you really need to do to understand the future of AMP is look at the amount of contributions that are being made to that GitHub repository. It fell off a fucking cliff. It is not being updated anymore. It is a dead standard. AMP does not have a future. So get rid of it. Put effort into improving Core Web Vitals instead.

Yeah.

And just forget about AMP. It's history, man.

Totally agree.

We won.

We won.

Um, how do you see commercial link building trying to get into news industry? Any editor's inbox can confirm that and how we evaluate the pure, uh, journalistic content. P rank benefit from links in news articles? I'm, I'm not sure if I understand that correctly, but are you saying that people are contacting you to have links in your articles?

Um, yeah, there's, there's two, two things here. I see this happen. First of all, some, uh, uh, publishers are being approached to, uh, engage in link building themselves cuz hey, links still matter. Uh, the answer to that is don't ever fucking do that. You should earn links with Never go chasing after links with any commercial partner. Just doesn't make sense and will get you penalized. Second, you will be approached all the time by people who want to place links on your website pointing to their third-party partners. And yeah, that can be commercially quite interesting cuz the bigger the publication, the more money you could charge for that. The answer to that is also don't do that. Google does tend to catch on to paid link placements, and if it's not labeled as a sponsored article or it's not tagged with sponsored content, Nofollow or sponsored, it can lead to manual penalties on your website and those can be quite painful.

Yeah.

I have seen some publishers, not many fortunately, but some publishers getting penalized for, for selling links. It tends to be fairly, uh, section specific. So if you do that in like say /sport, then your sport section will just not rank at all. And the rest of your website tends to be okay. Sometimes you get a sitewide penalty, in which case you're really screwed. And you need to really, you know, clean up your links to go back to that. So when it comes to link building for news publishers, be that either as a provider of links or as a buyer of links, the answer to both of those is don't do it. Just don't. Yeah. There's no value in it at all.

I remember a few years ago, I think, was it 800Flowers, uh, flower or flowers.com?

Yeah.

Wiped out.

The UK news sites, they got wiped out.

They got wiped out, didn't they? Yeah. They had a big, uh, digital PR campaign in the run-up. 'cause they wanted to be like number one for Valentine's Day flowers and stuff like that. And it had the exact opposite effect in that they just didn't rank for anything 'cause the entire website was deindexed. So yeah.

And news publishers got impacted as well. All the ones that were sharing them with no, like, you know, no follower, no sponsor or whatever. Uh, all of that.

Yeah.

Where does UGC fall into current news SEO, uh, best practices?

Mm-hmm.

I think you spoke about this and this question is from Julie.

Yeah, I wrote an article about that, didn't I? About, well, specifically about comments. I think there is value in user-generated comments. There's also value in user-generated content in other ways. For example, when it comes to product reviews and experiences that people have with products and services. I also think you need to be careful and you need to moderate that. I think unless you can have tight moderation controls on UGC, Um, or make your UGC non-indexable, uh, for Google. Um, some websites have very active comment sections and that's part of the appeal of, of, of people engaging with your website. But those comment sections are not that tightly moderated. They rely on people flagging content, uh, for example. And that's a very inefficient process in terms of maintaining the quality of, of content. In which case, by all means, keep your comments, but make them non-indexable. Let's just, you know, make it part of, of the user experience, but not the SEO experience. I think the only time you should feel entirely comfortable with exposing user-generated content to Google is when you have moderation over it. Editorial oversight, basically. Yeah. Where you can very explicitly say these comments get in, these ones do not. Uh, and that allows you to maintain a certain level of quality, certain level of language as well, because profanity is a negative signal in Google's ecosystem, believe it or not. Which is probably why I very rarely rank in Google. Um, so, um, but yeah, you have to be able to maintain minimum standards of quality if you want to expose it to Google. If you cannot do that, 'cause like Rachel just said in the chat, moderation at scale isn't cheap. This stuff is expensive. You have to hire a lot of moderators. Um, then you just, just make it non-indexable for Google and, and keep it out of the ecosystem because there's too many risks involved in terms of bad language and, and misinformation and other negative quality signals. And you don't want Google to see those.

Yeah, and, and 2 interesting things I know is one, if you moderate the comments, you are legally liable, uh, for these comments, right? So that's why some publishers either have no comments at all or, uh, they, they don't moderate the comments, right? And that's, that's a problem. The second thing, I have seen instances, Barry, where the comments were indexed and they change actually the, the focus of the article. So that, that specific article, it was like 400, 500 articles. It was very engaging. There were hundreds of comments on that article, and the way the comments went completely trailed what this article was about. And Google starts understanding this, this page or this content completely in a whole different way.

So, um, although Dan just made a comment in the chat, which is interesting because actually, come to think of it, this is an area where AI They're getting quite good at natural language understanding and can make comment moderation a bit easier and with less human oversight required. Uh, but again, tread carefully here because, uh, comments that are exposed to Google will be seen as part of the core content.

Yes.

On the webpage. So you have to feel confident about having that content exposed to Google in the first place. Yep.

Totally agree. Okay. Uh, be interested to hear, uh, your take on this as well. I always cite the New York Times achievement report. Would you recommend building one in addition to your XML, uh, sitemap as another crawling point? So HTML sitemaps and XML sitemaps.

Yeah, I like HTML sitemaps. I do. Um, and yes, I think there are some really, really good ones out there, not just, uh, New York Times, but other publishers as well that have good sitemaps that are like sorted by, uh, topics. Uh, like I see, I think it's The Sun actually, uh, that has a sitemap where they have, uh, the first page is like the trending topics and it's in the hot topics from like the last 24 hours. And then the rest is just alphabetically ordered list of all the topic pages that they have. I think they're very useful because they provide good crawl pathways for Google to very quickly find, uh, your topic pages, your deeper pages. Um, and also indicate the importance of those topic pages relative to how many clicks it takes from the homepage. Because from homepage to HTML sitemap to topic page can be 3 clicks, where sometimes it can be 5, 6, 7, 8 clicks if it's, it's a deeper hidden topic page. currently being covered by the website. So yeah, I am a fan of well-structured HTML sitemaps. Um, and because like I said, it helps with crawl efficiency and helps with signposting your most important topic pages.

Awesome. Thank you. Uh, let's take a question on Google Discover. I have seen a tendency in Google Discover where Google started to give some traffic to smaller niche publishers. Does it work in Top Stories the same way as in Discover? Are these ranking algorithms the same?

They are the same, but not the same. Um, let me explain that. Google relies on exactly the same measures of quality and relevancy in Discover, but there's different volumes associated with them. Discover uses engagement signals much more than like Top Stories. With Top Stories, Google wants to be relatively neutral and there's very limited personalization that goes on in, in Top Stories. Whereas in Discover, it's very heavily personalized, right? Google wants you to, to show you content with websites you engage with all the time. And so the core ranking algorithms are the same, but the volume buttons are basically inverted. Whereas Top Stories is neutral, Discover is highly personalized. So it becomes a slightly different ecosystem there. And that's where the engagement metrics also come into play, that if you have positive engagement with your website, you will get more Discover traffic because, you know, people engaging with your site on Discover means they will get more articles from your website in their Discover feed. But the essential algorithms are more or less the same because why reinvent the wheel? Google already is focusing heavily on trying to understand content and quality signals, and that applies to every part of their ecosystem where they, they surface content.

Yeah. Uh, this is like one of, uh, oldies, uh, but it's always good to talk about it every once in a while. Uh, path or a subdomain, would you recommend using subdomain or path for your So for example, uh, uh, xx.com/sport, uh, /news, or, uh, sports.xx.com/news.xx.com.

Uh, short answer, always use path, always use folders, and don't use subdomains. Uh, now this is context dependent, but it applies to like 99 times out of 100 scenarios. And that is because inclusion in Google's news-specific ecosystem is based on host path. Names. Hostnames is the complete domain name, like www.website.com. That's a hostname. And when a website is approved for Google News and especially top stories, that is dependent on that hostname. If you change hostname, like new subdomain sport.website.com, you will not automatically get inclusion in Google News. You have to re-earn it. And that's basically, um, because Google thinks a subdomain could be interpreted as an entirely separate website. And therefore doesn't automatically inherit any inclusion that is associated with the main domain as such. Um, also this is another aspect of, of Google's ranking systems, the way link value flows to websites. I have a very strong suspicion that an internal link from one page to another page within the same hostname passes more link value than a link to an external website. And a link from one hostname to another hostname is seen as a link to an external website. Yeah. Which means you lose the flow of link value through your website, which is generally why we see subfolders, paths, folders perform better in Google than subdomains because you inherit the domain ranking value of the main domain, including Google News inclusion, and you lose less link value through internal links. So yes, like Dan Moran saying in the chat, Rand Fishkin has been saying this for like decades. Yeah. This is not a new topic. We keep resurfacing it. Oh, there's no difference. The practical reality is there is a difference. Use subfolders. Do not use subdomains unless you absolutely have to. And if you absolutely have to, there are ways to mitigate it. If you have a subdomain that runs on an identical technology stack with identical design, identical top navigation, so that it basically behaves as if it's part of the main website with no other differences except that it's on a different subdomain, then you're fine. you might get away with using a subdomain. But even then, I think there's risks involved. So I would always say, try working with subfolders, with paths if you can. Don't try subdomains.

Thank you so much, Barry. With 3 minutes remaining, I was like, I just want to say thank you so much. I learned a few new things today as well. And I bet your audience have learned a lot as well. With the last, uh, 3 minutes, uh, a word of wisdom. What do you advise news SEOs and news organizations for 2024 and beyond? If you have one advice for news SEOs and one advice for news publishers, what would it be?

Um, that's a good one though. I think the best thing you can do as a publisher on the web, especially when it comes to news, is not put yourself in a, in a position where your fate as a publication can be decided by a third-party technology company. Uh, be that Google, be that Apple News, be that Facebook, be that Amazon, whoever it is. You need to have a loyal audience. You need to have an audience that wants to visit your website. I'll bring an example up from one of my own clientele. The Daily Mail have been hammered by Google algorithm updates again and again and again, for better or worse. You can agree with that or not. That's just the way it is. But they are thriving. And the reason they are thriving is, is because they get an awful lot of Google traffic. And you may think, well, what, what do you mean Google traffic? You almost never see them in top stories. Uh, and Google basically keeps demoting them in algorithm updates, but they get an awful lot of Google traffic. And nearly all of that is people typing in Daily Mail in the search bar. Their brand is so powerful. They, they don't really care what Google does to them anymore. They have, or are in the position of escaping the, the, the algorithmic treadmill because they bind their audience to them. They have built a brand strong enough that they can just bypass technology giants and get people to come directly to them. Now, not every website is capable of doing that, but I only need to point at a company I've, I've mentioned earlier today, 404 Media, 404media.com. They've been barely going for five, six months and already have a very loyal audience built around them because they have such amazing journalism on the technology industry. That's their focus, and I think that is the best way of building a media company. You will circumvent the the technology giants, get people to sign up to your newsletter, get people to sign up to your web registration paywall. Don't even ask them for money; just get them get to leave your personal details so you can personally engage with them. Build that loyal audience and, and try to avoid the, the algorithmic treadmill. And I think that's the best way of building a sustainable publication. And yes, that would make me as an SEO for news publishers entirely obsolete, but it would mean we'd have a very healthy news ecosystem. So I think overall that's definitely a net positive.

Barry, thank you again, uh, so much. Uh, that's a lot, uh, to, to learn from, and hopefully we will have the webinar, uh, available to you guys on YouTube, uh, soon. Uh, And once again, thank you for all our attendees. Don't forget the discount code, BAWEBINAR25, that we shared here. You have today and tomorrow to book for NEST of this year. Barry, thanks again, and hopefully we can see you again soon in one of the NewsDash webinars. Thank you.

Absolutely. Thanks everybody for attending and for the great questions as well. This was really enjoyable.

Thanks. Bye.