How Brie Moreau Studied 260k Search Results to Understand How Google Views Content
This week, SEO expert Brie Moreau doesn’t share his business results but, rather, the results of an incredibly in-depth study he did analyzing 260k search results to find an answer to a major question: what does Google want exactly?
Using SurferSEO data and a data scientist, he and his team set out to use a data-driven approach to understanding how Google evaluates content and they were able to draw many interesting conclusions, which he shares in this episode.
He talks about a wide range of topics, from the most important thing you can focus on right now to rank better to the advantages the “little guys” can have over authority sites.
The first half of the podcast is about the findings and the details of his research while the second half includes actionable recommendations.
Don’t miss this interview if you want a fresh perspective on SEO, content creation, and ranking in the SERPs.
Watch the Full Episode
Brie starts by sharing a bit of his story, which includes 18 years in the SEO industry, and creating his own digital marketing agency and conference, among other achievements.
At the moment, he’s focused on applying data science and machine learning to websites and, in this interview, he talks about a study of his published in Search Engine Journal
, which analyzes 260k search results and takes a very up-close look at how Google evaluates content.
Brie talks about how the study came about and then deep dives into the high-level findings.
First they talk about the connection he found between YMYL and health topics, and in that context Brie talks about optimizing using SurferSEO, the importance of topical authority and content clusters, and the idea of overoptimization.
He also talks about the massive importance of fresh, original content and how it has a positive correlation with higher rankings, and how repetitive content performs more poorly, especially after the Helpful Content Update.
When they talk about user intent, the conversation shifts to ERO – engagement rate optimization – and the importance of optimizing for engagement signals such as clicks, hovers, and scrolls.
Moving along, they discuss website speed and whether or not it’s an important ranking factor, and what surprising factor he and his team identify as being correlated with higher rankings.
In the second half of the interview, Brie shares his recommendations for enhancing topic coverage and how that involves content clusters and user experience, and what the real competitive advantage is in the current landscape.
They talk initially about topic coverage and how Brie goes about it at his agency, and the importance of answering questions efficiently and with depth.
Brie talks briefly about parasite SEO and site reputation abuse, and then shares tips related to site speed and engagement, such as checking for conflicting plugins and large ads which can slow down sites.
He concludes by talking about the value a data scientist can bring to an SEO agency and to an overall SEO strategy, and why people need to “flip the script” when it comes to trying to rank.
Links & Resources
Topics Brie Moreau Talks About
- His background in digital marketing
- His research and article for Search Engine Journal
- The importance of topical authority
- Surfer’s content score and overoptimization
- Topical authority and content clusters
- The importance of original content
- The role of engagement in rankings
- Topic coverage and content depth
- Parasite SEO and site reputation abuse
- Site speed and engagement
- Important takeaways
Transcript
Jared: We’re taking a data driven approach today. We’re going to talk data, and we’re going to leave
Brie: emotion out of it. Tell us more about the study. This we’ve never seen actually mentioned anywhere before. This is a novel finding that we’ve found. The better you do optimize for Surfer, the better your results will be in those niches.
Jared: Over optimized content, as it were. So what’s the balance there? Everything comes back to money, and so does the algorithm. And what that means is, What are you bringing to the table that Google has never seen before? That a search engine hasn’t seen before? What gain are you providing? The information gain is,
Brie: is the most important thing we can be doing right now.
Those two points right there are super fascinating. We have this really cool um, situation with one client. All they wanted was DRO and no SEO. We tripled their traffic in about six months just from A B testing this process.
Jared: One of the things you recommend, the first thing on my list is to enhance the topic coverage,
Brie: quality of the content that you write for each keyword that you’re going after needs to be,
Jared: how do people think about improving site speed in a day and age where they also need to improve engagement metrics kind of simultaneously?
Brie: Yeah. Like the, the, the thing is, is like you, you find that like if people are building their own WordPress sites, um, and. And the other, the thing about when you use like a dedicated WordPress, um, solute, a hosting solution is you.
Jared: All right. Welcome back to the Niche Pursuits podcast. My name is Jared Bauman.
Today we’re joined by Bri Rowe. Bri, welcome on board. Thanks for having me, Jared. It is good to have you here. We are, we are going to get, we’re going to get nerdy here today. We’re going to talk data. And, uh, I, I, I love topics like this and we get an entire podcast, but a topic that, uh, is really apropos, I think for the majority of listeners.
And we’re going to go through a study that you put together. You got published in search engine journal. It relates to Google. It relates to some updates and core updates. So we’re going to get into all the details before we do. Can you tell us a little bit about yourself?
Brie: Sure. I’ve been doing SEO for 18 years now.
Um, and when I started doing SEO, I started doing affiliate stuff before 2011, um, the good old days. And then when the first algorithm hit. I found myself soul searching and, um, and all that kind of stuff. Then I went to agency world, worked for some very large prestigious international multinational agencies.
Um, and love the technicality of the large site kind of work. But I hated the corporate life. Then I basically, um, quit my agency. Oh, it’s a, it’s a bit more of a story than that, but basically I was the first ever remote person in the company. I said, you have to hire me as a remote employee. And this was like nine or 10 years ago.
And they freaked out and they let me be remote in Australia. And then after that point, I said, I’m going to go to Bali now and I’m going to work from Bali. And they’re just like, no way, no way you can’t do that. And then I said, well, I am. And I gave them an ultimatum and they didn’t allow me to go to, to move to Bali as an employee.
So I moved to Bali and started my own digital marketing agency. So I have the longest running digital market, remote digital marketing agency out of Australia. Um, in that time, I also, um, set up a international digital marketing conference called DMSS. That basically, um, we had speakers like Ahrefs, SEMrush, all the big names.
We also had NASA as a headliner one year as well. That got killed by COVID. Um, we have a team of about 70 employees. We do everything from SEO, backlinking, um, web analytics, conversion rate optimization, and data science as well. And what, what we’ve been doing these days has been heavily focused on data science and machine learning, um, and really specializing in that area.
Cause we found that that’s, that’s, you know, once you start working on big sites, the next stage off that is actually to apply machine learning and data science to that process. And Um, we’re kind of innovating in that area a lot, but the data study that we’ll go through today is just kind of like a little snapshot of like, Hey, this is what you guys can see that we can do.
But the reality is the stuff that we’re doing within data science and machine learning is like, yeah, it’s, I don’t think it’s been done before
Jared: Well, why don’t you lead us into the, the study that we’re going to be going through today. Um, I’ll read the title from Search Engine Journal where the results were published.
260, 000 search results analyzed. Here’s how Google evaluates your content. And it’s, um, again, the subtitle is dive into analysis of the SERPs to understand how Google evaluates content with a recent helpful content update. Now we have talked at length about the HCU. We’ve talked at length about the subsequent March 2024 core update that you could say piggybacked off the HCU.
We have a lot of listeners because of what we focus on here, who have, uh, types of sites that were pretty heavily impacted globally in terms of like just site wide, uh, impacts for that. So this is a really interesting topic. Um, And I think data wise, we’re going to go through this. Now, a lot of people are gonna have emotion tied into what is done as well.
So I’ll just say at the outset, like we’re taking a data driven approach today. We’re going to talk data and we’re going to leave emotion out of it. That’s what Spencer and I do every Friday on the new section of the podcast. We rant and we rave about different things, but today we’re just going to look at it from a data standpoint.
Tell us more about the study. about how you came up with the numbers. Cause there’s so many interesting findings that we’re then gonna walk through and navigate each one topically speaking.
Brie: Yes, certainly Jared. Um, so the first thing is we collaborated with Eric Van Biscot from ClickStream. And so basically he was the guy that did a lot of the data studies for Brian Dean and Neil Patil back in the day.
So he was employed, is the data, uh, basically the, the, basically overseer of the data, a lot of the data studies for those guys. So he’s pretty much the, um. The granddaddy of, um, of SEO data study. So the, a lot of those, those Brian Dean ones you would have seen, um, that was actually Eric’s work. So we partnered up with him and, um, the reason why we partnered up with him is because basically he had an incomplete data study that he was doing for Surfer SEO.
And we kind of said, Hey, well, like, why don’t we come in with our data science team and basically finish off the study for you. So essentially, um, the study is technically Eric’s study in a lot of ways. And it’s not the way that we would have approached this study to be completely honest with you, but it’s what we had to work with.
And I’m getting exclusive access to surf ratios. Data is, um, incredibly valuable. So we basically overtook the projects and. So what happened was, um, what did Eric had done? And he’s, it basically, I didn’t, he’d got, um, to that, to, um, 263, 000 or whatever, uh, search in the results page pages. And then he also got, uh, I think like 8, 000, um, specific, like, um, keywords and terms and like segmented them into, um, banned, non banned commercial intent keywords.
So basically what we did in that process is we kind of over, uh, you know, We kind of like overlap those two databases together to start the, the data study.
Jared: Let’s get into some of the findings if we can, maybe we’ll have time to circle back on it. Otherwise, otherwise it’ll be a little bit too overwhelming for people.
I think you have to give them the, uh, you know, the good stuff at the beginning, keep them hooked, and then we’ll get into some of the nerdy. Uh, calculations. I do want to hear about it. I do want to hear about it. I do want to hear about it. But, um, uh, you know, I mean, I think, there’s just so much to unpack.
That’s what has me concerned. I’ve got 11 or 12 things I want to get to, and we only have an hour. So, um, we talked about this ahead of time. I broke it down topically. You thought that was a good idea, so let’s kind of topically navigate through it. If we have time, we’ll kind of get to nuance around about it.
We’re going to look at some of the findings. We’ll call it the high level findings that kind of came out of this, and we’ll go through it topically speaking. And I think the first one is the connection, whether you want to call it YMYL, which is your money, your life. And so that would extend beyond just health related categories into money, fitness, lifestyle potentially, right?
Like your money, your life can get into finance and fitness and health and all these things. But there’s wine whale and there’s fitness, uh, there’s health, sorry, and there’s the correlations you guys saw as it relates to the impact that this update and these updates had on health related topics. And non health related topics on YMYL pages versus non YMYL pages.
Let’s start there. Can you unpack what you guys uncovered with that?
Brie: Yeah, so essentially, there’s a positive correlation with, um, Surfer’s content score and YMYL pages. Meaning that since the core update that, um, that Surfer SEO’s content score is a higher predictor of success in those areas. And what that essentially means is the higher you, the better you do optimize for surfer, the better your results will be in those niches.
That’s, that’s the shorthand version. What does that mean in, in, in the real world? It means that basically the more that you have topical authority in that YMYL niche, that more, the higher predictor you have of success in that niche. Now what does topical authority mean? Topical authority obviously means, you know, like covering a topic, you know, broad and deep.
Right? So broad meaning like, let’s just say it’s the personal finance niche. You cover all the topics within personal finance and then you cover it comprehensively down as well. And one thing that I saw that I found quite interesting, and this isn’t like the issue that we have a lot of time with these studies.
Is that we’re obviously a lot of those big sites like Business Insider, entrepreneur. com, Forbes, and that they can just kind of cherry pick and, and put a page out and just outrank most people, which is what I’m sure a lot of people have been experiencing recently. Now, the way to, to get around that is to basically make sure that, cause you’ll find on the, on the very large authority sites, they don’t really have a topical cluster.
They terribly tend to have a, they just tend to write pages from the home URL and just, you know, spray them. So essentially. In order to, to, to, to kind of combat that and beat that, cause that, that was the old style, like you’d be able to have a smaller website and just, you know, put those, you know, slash URL and put that one keyword in there and then hope, and it used to rank right.
And now it’s more about having that cluster done in a really, really comprehensive way, which I’m sure your audience is aware of, but this is one advantage that, that, that the, the little guy can have over these authors.
Jared: When you talk about the content score from Surfer SEO, is that, is that I mean, I, I’ve used it, but for the general sake of the audience, is that basically focused on optimization?
And the second question of that would be, and maybe navigate through this, a lot of people surmised that, uh, the HCU and subsequent March core update actually came after over optimized content as it were. So what’s the balance there? Was there any sort of balancing act you saw in this study?
Brie: Yeah, of course.
Over optimized content, you know, insert keyword here, keyword here, keyword here, keyword here. That would obviously send alarm signals to Google. Now, on the other side of that equation, if you, if you, the, what, what we’ve done regardless of this, this data study, what we’ve done is we’ve taken Bert’s, um, Google’s, Bert LLM, Google’s basically large language model.
We’ve, we’ve basically created an embedding space there. And this is pretty much what Surfer SEO and these other correlational SEO softwares do. They basically sit there and say, but it basically gives you the map. They say, these are all the clusters around these specific keywords. And then you take your website, Surfer or these other softwares.
They take the, your keywords on your page and all this stuff, and then put it into that, that, that embedding space. And they see, okay, how much of this cluster, which points in this cluster, this, does this query fulfill? Now, if you, if you basically fulfill the cluster and make it a full cluster, And there’s no such thing as over optimization.
You’re basically completing the jigsaw puzzle for lack of a better way of doing it. Now, when you look at it, um, everything comes back to money and so does the algorithm and what that means is it comes back to information retrieval costs. And so if you, if for Google, you know, there’s billions of searches happening a second, you know, this is so much volume in searches and Google needs to find the quickest and most relevant search in, you know, a 10th of a second.
And in order to do that, it has to basically cluster all the content into different pieces. And the stronger your cluster is, and the stronger that signal is, the quicker it is for Google to go, this is the right cluster to retrieve the result. Does that answer the question?
Jared: It does. Yeah, that does. And I think it does draw a line as well into maybe segmenting or segueing into the next one because this is specifically for health and or YMYL content.
And the results were, I’d say a lot, not a lot different, but there were, there were varying degrees of results when you moved away from that, you know, and so another detailed finding, another kind of intricacy of what you guys saw was that on page rankings and this general importance of content coverage.
The inclusion of keywords, the inclusion of phrases, and that was beyond just in the YMYL space, YMYL space. So maybe talk about the differences there. Why the segmentation of health versus all the other topics that people could cover and why, how does that relate to the on page factors that you surface?
Brie: Yeah, because essentially, like the way that I like to explain this is basically, um, there’s as many algorithms, Google algorithms, as there is keywords. Because every single keyword has its own user engagement metrics, its own algorithm in there. So we, we, we, we’re doing a whole bunch of other stuff that it’s going to take too long to explain and we’re not specifically relevant to this question, but essentially the, the more that you, you, you segment, um, you know, like this is take an example, like a one niche, you know, you might only have to write like 2000 words of content to rank in the top 10, another niche might be 5, 000, the other one might be.
Like keywords, sorry, not even, you know, within the same niche, even some of it might even be a hundred words because it’s basically, you know, a buy in 10 keywords. So you really, you really need to segment as much as possible to understand how best that niche operates, because if you’re, if you’re just blindly, and this is what tools like Surfer do really well, it kind of gives you the paper by numbers.
It tells you in this specific keyword, if you want to rank. This is the sweet spot. So this, this is why we segmented everything like that. But from an overall perspective, like beyond what surf has done, like we, we segment things into obviously page load speed and other variables as well, because essentially these are kind of like other ranking factors that, that lead to that process too.
So that’s why we’ve kind of segmented this way. One interesting founding that I, um, that we, that we came across in this study, that I don’t think it’s, it’s, uh, spelled out very well in the actual study. Is there specific keyword phrases that, um, that kind of signify to Google or to, to Google or to Surfer that, um, this content is, is poor quality.
And I would say that I’m not sure if this is the right way to put it, but essentially if face is specific phrase repeated across the web, let’s just say it’s a specific phrase on like, you know, um, you know, eating chocolate mousse makes you fat and a specific way that that’s said. If that kind of phrase specifically or very close to it is repeated across the web, which pick, you know, writers is kind of repeat other content.
It’s still a hundred percent unique, but that phrase might be identical. This was a predictor of negative correlation. So the more, the more unique the content is, the more outside, the more fresh and the more variants from what everyone else is, you know, cause you look at a writer, they’ll just see the top 10 results, pick the two or three articles and just rewrite that in their own words.
Thanks. And the more kind of away you get from that, we find that that’s a predictor of increased, um, increased value of rankings.
Jared: This might be a good time to bring it up. You tell me, going a little off script here, but there’s a concept, uh, that’s been, you know, routinely thrown around as it relates to different core updates, especially hopeful content update, as this idea of information gain, like, Hey, you have an article and whether it’s a, uh, an affiliate article, whether it’s an informational article, whether it’s a, a comparison, whatever type of article it is.
The idea of information gains, like, says basically what are you bringing to the table that Google has never seen before, that a search engine hasn’t seen before, what gain are you providing? Um, I didn’t see that directly mentioned in the article, but I’m, I’m hearing so many threads in what you started to talk about in information gain.
How does that correlate with what you’re talking about?
Brie: We had a negative correlation there, meaning that, um, it’s not specifically mentioned because Eric, Eric wrote the article, but essentially what we saw was if it wasn’t that way, there was a negative correlation, meaning that if it was that way, there would be a positive correlation.
So information gain is, is the most important thing that you can be doing right now. And like what we’re telling our clients is basically user generated content. Um, like a whole bunch of, and you will see that since the helpful content updates that we work in the travel space. I mean, we work in different niches, but one of them is travel space.
And what we found since Healthful Content is, um, you know, this is a very large, you know, multinational like tour operator. And what we’ve seen in that space is, you know, you know, Joe’s, um, blog about Istanbul, who’s a local in Istanbul doing his tours is coming up significantly over these, you know, you know, multinational tours, because he’s got that localized content with the pictures that are all his pictures.
And he’s like, you know, you really need to to this little, you know, kebab place here. And, and this is the kind of content that Google is rewarding more of. It’s fascinating.
Jared: You know, what’s interesting is we talk about your money, your life, and then we transition into this information game. There’s a lot that people can take from that.
If you don’t want to get into the data nerd, like you don’t nerd out on it. Those two points right there are super fascinating from just a very high level, just a very approachable way to look at something like this.
Brie: Right. And I’ll give you another one that, um, that, uh, that a friend of mine who runs a very large site has done.
So he works in, um, kind of like e com, but it’s kind of like, um, it’s basically an e com site. But what he’s done is he basically pays people to unbox the products. And then he gets a refund for the product. So he pays someone to be a professional unboxer. So they review, they review products, right? And then someone just gets all these products delivered to their house.
And then they just take the photos of them unboxing the whole thing. And then they put it back together again, and then they get a refund on it. And they do this with thousands of products. And just having that unboxing process inside the content has worked really, really well for him.
Jared: Fascinating. That is very, okay.
Well, we might come back to that if we have some time, I have a couple more questions I could go off on, but I want to make sure we get to everything. Um, another thing that you surfaced is user intent. Um, and this has long been discussed, but I’m curious to hear more from you on your findings. You specifically seem to segment informational content in this, in this part.
Um, and, um, and talk about how optimization, like it, it really, you, you guys found that it needed to be better optimized for SEO compared to say buyer intent content. And I thought that was interesting. I wanted to hear you maybe share more about your findings there.
Brie: Yeah. So, um, essentially like I’m like the raw binary data is like the raw binary data.
Like what I’m giving you is my interpretation of that data. Perfect. That makes sense. Yeah. Yeah. Right. That’s right. That’s zero in the one positive, negative. Um, yeah, from a, from a buyer intent perspective, it’s, it’s incredibly, um, how do, how do, how do I put this? Like you use from a user intent experience.
The way to look at it is that because Google’s processing power has increased significantly. Um, you, they’re able to use more user engagement signals inside the algorithm now, and because they’re able to use more user engagement signals, they are actually using them a lot more. And what you’ll find if you’re at a highly competitive niche, what you’ll find is that we’ll, we’ll, we have clients that have very large backlinking budgets, like, you know, 50, 000 a month plus, you know, and, and going up from there as well.
And so. What we found is that like, no matter how good the content is, no, how aggressive we might be, have a velocity of like three times of our competitors for link for a specific, you know, for a specific keyword term and no matter how much we do, we still can’t shift the top three positions and you’ll, you will have noticed that intuitively as well.
Like you were like, well, this one’s never going to move. Yep. Doesn’t matter how many links I build and how much content I do around it. It’s just never going to move. And so what we’re telling clients now, okay, cool. He’s spending 50, 000 a month. We’ve, we’ve hit a really aggressive boot. Clearly. Should own, you know, the top position, but the top position in those highly competitive keyword terms comes back to user engagement metrics.
So what we’ve been doing, we’ve actually calling it ERO, engagement rate optimization. So we do conversion rate optimization, but we focus only on engagement rate signals. So we’re doing CRO, like, like in the sense of like, it’s A B testing, but instead of like, you know, the goal being a transaction, the goal is like keeping a person on that page.
How do we do that? You know, making sure they click through to another page, making sure that they, they scroll down, you know, making sure all those other kinds of like, um, engagement metric signals work really well. And the really cool, we have this really cool, um, situation with one client. We didn’t even, they, all they wanted was ERO and no SEO.
Not sure why, but we were like, Cool. That’s what you want to do. You don’t want to do SEO. We don’t want to optimize pages. We tripled their traffic in about six months just with, from A B testing this process. And this was quite a large e commerce site too. So like, and that, that’s just a once off that we had that unique situation, we’re able to do that.
But from our perspective, it validates that if you just optimize for engagement signals, you will perform well, especially, and most importantly, especially in high traffic keyword terms, where they have enough data to do A B testing. Just for someone
Jared: listening, like, what are some things for people to look at maybe when they’re evaluating the SERPs and they’re looking for signs that a page is going to have higher engagement metrics than others because, you know, we don’t have data on that unless you own the site, right, unless you have, uh, access to the, to the, the analytics or, you know, those sorts of things.
So what are some things people can maybe learn from this? Because I wouldn’t say this is a new topic, but for a lot of people seeing this show up and data reports and studies and. Obviously we had the Google leaks and we saw a lot of information there about it and stuff. So how can people maybe analyze any tips you have?
Brie: Yeah. So essentially, I mean, I, I would just talk about the site that you own. It’s probably the best way to look at it. Um, and the way to look at it, I mean, first thing is like, you know, the sniff test and make sure that, you know, your site is actually optimized. Well, you know, like, let’s be honest. Right.
Yeah. And the reality is like, the reason why we started doing this process too, is because we do, we do traditional CRO and we’re like, man, you know, 95 percent of the traffic is going to the blog. We’re wasting all these opportunities to test in this area. And that’s, you know, and that’s what people need to kind of look at.
They’re like most, most of the traffic goes to the blog. So why not test on that? And why, you know, if you just get a 0. 1 percent of an increase in conversion there, that will filter through to your site, you know, overall, you know, people just kind of look at it as like a dead traffic, you know, but essentially the way, the way that we approach this process is obviously we set everything up properly or the tagging or the tracking.
But what we’re looking for is a couple of different things. We’re looking for kind of scroll depth. So basically clicks, hovers, and scrolls, which is, which is, we know what the Google quality raters are kind of looking for clicks, hovers, scrolls. And so what we do, let’s just say there’s an article on seven tips on how to lose weight.
We do, we set up heat map trapping for that, for that page, but there’s also other softwares will sit there and show you hover time on a specific area. Yeah. So let’s just say that you have seven tips on how to lose weight. Now you put the heat map trapping on there and you’ll see that people will look at, you know, uh, tip two.
And tip four and tip six, a lot longer than the other tips. So what do we do? It sounds really simple, but not many people are doing it. We just put those three tips up to the top and we expand those tips out. And consistently you will get a higher engagement, higher engagement on that. Besides all the usual stuff, which we, you know, which obviously most people’s headers are too big or the, the, you know, the advertising block is way too big and things like that above the fold.
You know, you just need to kind of get them past the above the fold. But that’s a really, really simple, actionable thing that people can do. It’s great. And that’s not hard. You don’t need to be a CRO specialist to do that. It’s like put heat map tracker software in there, see where people are actually spending time on that page, move them to the top.
And that’s, that’s, that comes back to the user intent. That’s pure user intent, right? Because you know, there’s someone typing in how to lose weight. And those three tips are the ones that they like. You’re basically creating the best positive feedback loop for Google you can possibly ever
Jared: I just want to double down on what you said.
Like so much can even be learned just by having an objective look at your page and understanding how it, how it is for a reader and not for you as a content creator, because you can get so wrapped up in the way you present your content, following a system. The way you thought about it, but you’re so wrapped up in the research you did, the writing you did and all that.
But looking at it from that third party perspective, if you can, to look at it and how a reader would engage with it.
Brie: And Tata doesn’t lie, man. You know, yeah, you know, you, you’ve, you might think he might have had the most beautiful baby in the world, but Tata doesn’t lie, you know, just to bring it back to, you know, there’s often a common thing where you, and we’ve started to see this pattern as well.
Cause we really are looking at everything quite differently. We’re not. We don’t, we don’t look at things from Ahrefs, um, or like, we don’t just sit in the, the SERP results and go, all right, this is high DR, this is low DR. We actually go and analyze each site. An example of, of, of what to do in this situation.
You’re like, all right, everything’s a DR 50 plus in this niche. Right. And then you have one or two DR 20s. It’s like, why are those two DR 20s in here? Why? You’re like, oh yeah, they’re just testing, you know, the, some sites and try to create some, you know, variance in the search engine results. What you’ll find a lot of the time, those smaller sites offer something to the user that other, other sites don’t.
And a really good example of that is we were working in a really highly competitive supplement niche and we had sites of DR60 and a plus in, in that niche and like really, you know, strong sites being well established. And there was this one DR20 something site in there. And we’re like, what is that doing there?
We click on it and it was a pink, pink website sites for women’s supplements. What does that tell you? That’s interesting. At least half the people searching for the traffic are women, and half the people, they want to see that pink sign. And that pink sign’s a good wedding
Jared: for them. I was going to say, that appeals to a large enough group of people, they make themselves into the top 10 and stay there because of those metrics.
Brie: Exactly. And so when, when we do look at that, you know, when we’re looking for that variance and we’re like, okay, obviously as, as most of your, um, viewers, listeners, I don’t know what the definition here is. Um, but all of the above. You know, so when we do see those like anomalies, like it’s obviously we’re analyzing, okay, there’s top 10 results.
There’s, you know, four authority sites. We can’t take over. These are the six we can take over. Why, how can we take them over? So when we’re looking at that, we’re not just looking at it from a content length and backlinks to page. We’re looking at it from an engagement metric.
Jared: Let’s
Brie: move on. If we
Jared: can, got a long list here, got to get through it all.
Let’s keep going. This is good. We could probably do a podcast on each of these topics. The next one, super interesting. This is maybe the one I thought I spent the most time on. I’ll just leave it there. Website speed. Website speed and especially a couple of factors. Time to first byte. HTML file size, some things that really are interesting to nerd out on and talk about.
So, hey, look, website speed has been something Google has dangled before our eyes for years. It’s like, get your site faster. We’re going to start penalizing you for it. Never did. Never did. There still isn’t perceived to be an outright penalty for a slow site, but there certainly does. And again, I’m repeating what people talk about in the industry.
There certainly does appear to be like a tiebreaker effect. Is that no longer the case? Is it no longer a fast site? Is a tiebreaker to win? And how important did you guys find specific aspects of site speed in terms of rankings?
Brie: I, I would disagree that um, the website speed has never been a factor. It’s always been a factor.
Um, like that, that’s pretty much a given. Um, and also, for, also think about the Google crawler, right? The Google crawler has to go and crawl, you know, trillions of pages daily. Like, it’s not going to sit there and wait whilst your site takes, you know, you know, seven seconds to load on a WordPress thing, right?
So that’s, that’s the first thing. So I, like, I think it’s always been kind of, um, a big thing. Um, and I think that like, I do find that with technical SEO, a lot of people get stuck on page speed and think that if they just solve the page speed, that, that, that everything will just solve from that. Because, you know, it’s, it’s a very quantifiable metric, you know, your website is this slow, if you increase it, they seem to think that it magically solves everything, which we know it doesn’t.
But at the end of the day, as far as like, you know, a healthy diet, You know, you need to make sure that you have that in your diet to, you’re not going to just, you know, you’re not going to like win the SERPs because you have like a, a one, a one pixel, uh, website that says, you know, the keyword or whatever, you know, that’s, that’s, that’s not how it works, but definitely, definitely positively correlates in all of this stuff.
So I definitely think it’s worth important, important to do, but I do find a lot of technical SEOs and people obsess over this metric more than other metrics, like for us, it’s. Probably a tertiary or, or what the other one is after that, but like it’s not the primary metric that we go after. Now the other thing that you men that we, that you mentioned there about HTML five, five file size.
This we’ve never seen actually mentioned anywhere before, and this is a novel finding that we’ve found and that basically the size of the file is been positive. The larger the file size, the positive correlation with increased rankings. Now why is that? We don’t specifically know, but we can. I suppose, like assume it’s because it’s the breadth and depth of the content and all the kind of like media that’s associated with that thing, not having a buggy, you know, inefficient piece of code, but actually the more the elements inside there, like, you know, lots of images, lots of media and things like that.
So this is something that we found that was quite interesting. Um, and that, that, I don’t think that’s actually ever been found before.
Jared: Now that increased, the correlation increased substantially in the March core update. And, um, and you said in the article, this matches squarely with Google’s statement that user experience is its priority for helpful content update.
Uh, for the helpful content, sorry, time to first byte is the most important factor as it was five years ago, but HTML file size was kind of new. So, I mean, is it just because helpful content is breadth and depth? And so it, how, how does that play into site speed? Yeah.
Brie: I mean, I think for me, it says it all in the name, helpful content.
You know, like what’s helpful content. Is it like, you know, 800 words of content on a page or is it someone unboxing, you know, the thing and having the detail and all that process? Like it really, for me, it comes back to that media richness or I suppose web experience richness. That that’s, that’s how I would interpret this.
It doesn’t mean you go out there and just whack a whole bunch of HTML on your side and do that process. It means the more features you have on your page, the more HTML code you have. The, the, the, the, the, the more that your page, like a simple WordPress site, that’s like a, a themed WordPress site that everyone has the same template.
That’s not going to perform well in these days. You know, you need to have that dynamic content, whether it’s animation, specifically animations, but like there’s JavaScript, but, um, but you understand, like it needs to have that, that fully textured experience. And so, yeah, I, I, I don’t know how, why that’s like that with thing, but that’s, that’s a unique, interesting.
Okay.
Jared: Um, I mean, maybe we can just navigate into the second half of the conversation into recommendations, right? It’s always difficult to make blanket recommendations, I find, from stuff like this, but certainly using the data to help suggest people look into certain areas, go down certain paths, potentially consider certain things for, for their site.
Um, I mean, let’s get into recommendations. I’ll just start at the top from the things you guys sort of surfaced. A lot of this might be Be playing back on things we’ve already talked about, but maybe we’ll try to put a bow on it so people can know where to go with it, right? Like first half of the podcast was really about the findings and the details of those findings, maybe second half is really, where do we go?
So I think the first thing that you guys recommended, or one of the things you recommended, the first thing on my list is to enhance the topic coverage. Um, where do people go with that? How do they analyze that? And how do they, and by the way, uh, a second part of that question is. What do you recommend doing in areas where they can’t enhance topic coverage, either because of budget, because of time, because of knowledge, expertise, gaps, stuff like that?
Brie: That’s a really good question. Budget’s always an issue, right?
Jared: We both run agencies, that’s always one of the first things that come to mind, right? But even budget for someone running their own site and needing to hire other people to help them with these kinds of things, topic coverage, it’s expensive.
Brie: Yeah, so essentially the way, the way that we approach SEO is, is very different from many other agencies and, um, what, what I find is that, um, people are very divorced from the search engine results page. So they basically go into Ahrefs or whatever their tool is and they extract the keyword list and go low keyword difficulty, yellow or green, and then do this thing and they’re like, all right, content writing team, take this information and rate the content.
You know, and this is what we see. Right. Um, and this is, this used to work. And so now what we do, so we, we challenge so many assumptions in this space and people, when they hear topic coverage, you’re like, I need to write more content. I need to write more content. And, and this is not the case. The case is your, the quality of the content that you write for each keyword that you’re going after needs to be like, we kind of have like an oneness with the keyword that we’re going for.
Like our, our head of SEO has been doing. So for more than 15 years, and we, we, we pick, we pick, we’ve got sites ranking with more than 10 million a month traffic with 20 pages and that’s it. But the reality is those 20 pages cover the topic well, and it doesn’t necessarily mean that it’s like, you know, they’re all like 15, 000 word letter, you know, essays.
It just means that we understand what the search engine, the results pages are showing us, and we can kind of fulfill the queries within that. And so this comes back to the kind of like whole topical map area. You want to be as efficient to Google as possible. So creating a topical map and kind of like filling out those kind of like content clusters in a specific way that does, there’s two things.
There’s, there’s the BERT LLM where you want to make sure you kind of hit all the kind of points that need to be hit within a cluster, then there’s the user experience process. So the hitting the points is quite easy to do it. And someone that can just kind of like be very binary can, can hit those points quite well.
The second part is more nuanced. And the second part is like, okay. What are people looking for here? And this coming back, that example of like, you know, seven tips to lose weight, which are the most important tips that we see? Like, let’s just say there’s 10, a list of like reviews, right? And you, you have a, you have 10 review.
Um, let’s just say it’s, um, best vacuum cleaners of 2023. Right. And you, what we, what we do is we can basically analyze all those 10, the top 10 list, list results, and actually put each specific vacuum cleaner that’s actually listed in each one. See which ones are, um, The same across all of them and put them first.
So we’re really dissecting the search engine results. We’re not just going, all right, we need to write about 10 vacuum cleaners. Let’s pick 10 vacuum cleaners. Let’s go, all right, out of those results, where, which ones, which ones have the same vacuum cleaners? Maybe there’s a correlation there between these ones because there’s 50 pages exactly the same, but these are the 10 that seem to work.
Maybe people tend to hover over that area more. So we really try to go the extra detail because essentially we’re not interested in, in writing. Everyone has the same. These correlational tools like Surfer and these, they’re not a longer a competitive advantage. Eight years ago, they were very much a competitive advantage.
But not a competitive advantage. Everyone has the same tool. What’s the competitive advantage now? It’s actually being tangible in the search engine results. And another thing that we’ve done with a client recently, cause we just get, I just get tired of clients coming to me and saying, Oh yeah, we’re writing four blog content pieces a week.
We’re trying to rank for best vacuum cleaners and writing all this content. And I’m just like, does it work? And I said, look, show me all the, the, your content strategy for the past year. Yeah. And, and so we, we, we, we do a lot of backlinking that’s, that’s really where the strength is, but this is different to this, this topic, but essentially we’re like, we said, basically give us all the topic contents you wrote about vacuum cleaners, you know, fridges, explain how many pieces of content that you wrote and let’s do a baseline metric for the past 12 months, if that actually went up or not.
Lo and behold, it didn’t because essentially you’re creating a whole, and that’s just one self with a client, but I was just trying to like, they didn’t, they weren’t able to get it. And I was like, look, let’s, let’s do the data process on this, right? You’ve written like 12 pieces of content about vacuum cleaners in the past year, but they’ve all been shallow pieces that don’t have any real value to users.
And you haven’t gone up in rankings, writing all content is not going to solve your problem. 12 pieces or, you know, 36 pieces is not going to solve this problem. So if we’re really trying to like hammer, hammer home this process of how to get that depth there.
Jared: So if we look at content depth then as more than just writing more content, would you say that’s a more holistic way to look at what topic coverage means or are there any additional features that people can think on?
Brie: So bringing it back to the BERT large language model, which is everything for this whole algorithm is based on, or this part of the algorithm is based on, so we have a client in the casino niche, right? And in that niche, we had to go down to the level of how playing cards are manufactured, you know, what materials are they used in playing cards and all these different things, because we really have to understand all the topical maps within that area.
Okay. And that, that, that’s irrelevant from a user experience. And that, and that, that specific keyword would not show up in, um, in IHS because no one’s like typing it. How is it manufactured? But from the LLM perspective, it needs to be there.
Jared: Second thing, uh, high level thing. And we haven’t really talked about it.
We’ve kind of talked about it without talking about it. EEAT, um, specifically called out for health, but really since you guys were talking about it across the board. Um, you know, it’s not a new topic for a lot of people, but, um, what came out in terms of very tactical recommendations from this study as it related to that topic?
Brie: So, you know, there was a Parasite SEO, um, update that kind of hit like in May. Um, we’ve been working with a lot of clients at the moment, specifically on, with, with there’s a, there’s a, there has been an absolute annihilation in the Parasite SEO space or like on this column, new sites, uh, space in the past, like three months.
Um, and it hasn’t really been that public because, um, Most people kind of have blocked sites. They’re not really across the whole, that whole parasite SEO phase, even if they are buying parasite SEO sites or actually buying links from these places, we have seen like it’s, Google has been incredibly aggressive in this area.
We’ve worked with a lot of clients come to us and they’ve lost millions of dollars of revenue daily and they come to us and we’ve had to kind of do reconsideration requests and all that kind of stuff. The one thing that I can tell you is that we, we have in a lot of it is to do with site reputation abuse, which is what you know it as, and part of the issue in that process.
Um, and this is a little bit off topic and probably a little bit higher level than most of your audience, but just to kind of get the concept in the head. Um, basically the site reputation abuse comes from, um, external writers writing on and then leveraging your platform to, to rank higher. Now, in one of these situations, we had like.
We’re, we’re working with some of the biggest parasite sites in the world, to be honest with you. Um, and the sites that you know that have actually been hit, um, that you, that you would know the sites. Um, and one of the ways around that process was, um, that basically we, the client basically, uh, took a, a, a rider, a parasite SEO, like very large news site client.
And the client and the Parasite SEO site got hit because they were very, very aggressive with what they were doing. Um, not our recommendations, but it worked really well for a very long time. And basically what our client did was they basically employed one of the writers from the Parasite SEO site and basically said, we actually, they basically pay the Parasite SEO site.
a salary for that employee. And so they basically took their employee, put it across the parasite, the parasite SEO site, like a full time employee. And now that person now works for the parasite SEO site and says, Hey, Google, we actually work for this site. So anything that you said before is incorrect because this person actually works for that, for this, for this, for this thing.
So that’s not really relevant to most people out there, but for some people that are listening, that are having this issue, this is one way to get around that process. Now, to bring it back, what we’re doing with these, um, with these clients is they’re basically having to go through and do massive, um, author audits.
And a lot of these sites will have so many freelancers for the, it’s like the backlink, um, It’s like, you know, backlinks, any backlink work, you know, you know, before the first algorithm update. And it’s the same thing with autoscan, any author work before and now it doesn’t work anymore. So the way that we approach authors is like a backlink.
We look at the author profile. We look to see what, you know, we type in, you know, author, you know, search operators into Google and see what, what, what they rank for and what articles they rank for. They rank for crypto. They’re not going to be writing about payday loans. You know, or, you know, or cannabis, all these other things.
So we basically make sure that the backlink or the author is relevant to the topic. And the second thing that we do as well, it was, you make sure that every, well, we’re trying to include that with clients is making sure that there’s a chief editor across everything, which is really easy. It’s not hard to do.
And if you have a lot of freelancers working on your site, then you just make sure that there’s an editor that’s actually named is. Attached to all this stuff, all the usual signals that you know about authorship. Perfect.
Jared: Um, another topic that you brought up in terms of recommendations, and we’ve touched on it, but we touched on it from more of the theoretical, that is site speed.
Improving site speed is important. Like how do people navigate the nuances of improving site speed? And I’ll say, Because you brought it up, I think it’s a fair point to say, like, it’s not just about having a WordPress page with a bunch of content on it, because that’s not going to increase readers happiness, it’s not going to increase their engagement.
So you’re having to add additional elements, whether it’s as basic as, say, YouTube embeds, or more complicated, like, Features that engage them, animation, GIFs, quizzes, these interactive elements, but that can have a drain on site speed. So maybe from a high level, like how do people think about improving site speed in a day and age where they also need to improve engagement metrics kind of simultaneously?
Brie: Yeah, I mean, essentially that really, I mean, essentially it comes back to, um, to, um, to servers as well, right? It comes back to server and time to first byte and all this stuff. So obviously. The simplest solution for that is obviously a dedicated word. If you’re doing WordPress, having a dedicated WordPress, um, we use wp, um, WPX hosting, um, they’re quite good at cost effective.
Other people use WP Engine. That would be the first place that I would start. It’d be like, you know, what’s your, what, what is your server? Is it WordPress? Uh, is know dedicated? Is it optimized WordPress? Yes or no? If it isn’t, then move across to WP h or WPX hosting and WPS hosting, um, doesn’t have those traffic limitations that WP Engine has.
That would be the, probably the first place to look. Then it’s like, it’s just the usual stuff. It’s like making sure that there’s not a whole bunch of unnecessary code on the site that’s, that’s slowing everything down. Um, and yeah, like the, the, the thing is, is like you, you find that like if people are building their own WordPress sites, um, and the other, the thing about when you use like a dedicated WordPress, um, solute, uh, hosting solution is you can’t use all, um, All the plugins that you want because they have their own versions of those plugins too.
So it’s kind of like a one stop shop. Yeah. You just kind of bring it across and you might’ve had some like some optimization, uh, speed optimization software on, on your WordPress site, but then you can’t put them on because they basically, um, they have their own variations of that. So I would say that’s the quickest and most effective way beyond that.
It’s basically making sure that there’s not, the problem is there’s a lot of people that are building these WordPress sites and not developers. And so they’re basically have a lot of the issue that we see a lot of is conflicting, um, plugins that are out of date and things like that. And the unnecessary for that specific page.
The other issue that you’ll see a lot of, um, is unnecessary code from a tracking perspective. Um, and specifically if they’re running display ads, you’ll find that a lot of these ads, um, a lot of these ads are, uh, uh, uh, codes is super clunky. Like some of them, you know, like Mediavine is like super slow.
Like you’ll find that some of these ad blocks are really, really, really slow. And another, just to riff off that a little bit as well, um, with all these updates that happened in the past, like, you know, 18 months. Um, obviously advertising space on a page like Google’s guidelines, I think a 30 percent of the, the page can only be, can be, um, can have advertising on it.
And, um, and what we found was basically like, you know, you go on a site these days, you go out and search a journal for God’s sakes. Like that, it’s just like salt. It’s like, you know, you’re lucky to find the, like, you know, the real original content on there. Right. Like, I, I don’t know how, And what you’ll find with like Mediavine and I’m not saying Mediavine specifically, but what you’ll find a lot of the time is you’ll set the aggressiveness or the amount of display area that your ads will take on a specific, um, you might go, Oh, I only said 30 percent because, um, because that’s the safe limit for Google.
Right. And they might even tell you that, but on the back end, if you actually render that out across, um, you know, different, um, you know, uh, devices and, and browsers. It’s might be taking 45 percent of that space. And that’s so that the case for me with search engine journal. So people need to be super aware that even if they click the 30 percent button on their, um, their display advertising software, that it’s not necessarily the case and you’ll, you’ll see it anyway.
You’re going to find, you’ll see like a scroll forever just to get, you know, paragraph of content. So this is a really big thing and that’s not necessarily read to site speed, but it kind of is as well because it does come back to user engagement
Jared: from a high level. You know, I think, like, having mastermind and collaborated on this, uh, on this testing, this, this article, like, are there any other high level takeaways you would want people to know or consider as they look at this data?
You know, because for a lot of people, the HCU and the March core update had a lot of emotion tied to it. So is there anything else that the data says that we didn’t discuss that you really think is important that people walk away kind of as we close here? Thank you.
Brie: Um, no, I, I, I, I would actually like to say something else, you know, and I think that that is, um, people need, like this process that he went through, um, you know, we’ve, we’ve had a data scientist on the team for about a year now.
The process we went through with analyzing this data has made us infinitely better SEOs because we’ve really had to like look at multiple variables and just kind of think about it. And the output might be, you know, you know, a publication on search engine journal. But the real output and the real gain for us is really understanding how the data saying the mechanics and the way that I envision the future is that every single SEO company or media, whatever, if they’re working in this space, they have a data scientist in their company specifically tracks the algorithm for their specific keyword.
And this study, we kind of like inherited, as I said, it wasn’t our study to begin with. So we kind of had to work with a framework that was already there, but I would sit there and say that if you created a, um, a kind of like a framework for a data study within your specific niche, every time there’s an update, you just extract the data out and then run it again, you know, and, and, and try to like, what I’m encouraging people to do is to take a, to kind of like be in charge of your own algorithm, if that makes sense.
Yeah. Because essentially it’s so nuanced for every query and every space and every niche. Yeah. And that’s the thing, right? Like you really, really want to understand it. And the dot is that.
Jared: Yeah. I mean, you’re, you’re highlighting on the, the health topic specifically, but really, I mean, to your point, like you could highlight every segment, every sector, every, and then you get into localization and you get into all these factors where the algorithm isn’t just the algorithm.
It’s more nuanced.
Brie: I really encourage people to get into this space because this is the future. Yeah. This is really what it comes down to. And like, you know, like I, I want to collaborate with more people in this space and, um, but at the end of the day, like, it’s just like, we need to, we need to flip the script and we need to own more of what happens in this space as opposed to freaking out and then, you know, trying to find the thing on search engine journal or search engine round table.
Like, what are we doing here? What do we do here? It’s like, you know, like there’s a certain empowerment.
Jared: Well, Brie, thank you for the study. Thank you for coming to the podcast to talk about the study. Um, where can people follow along with what you’re doing?
Brie: I don’t know, I don’t, I don’t, I, um, we’ll have a website.
Uh, we don’t act, I didn’t actually publish anything to be honest with you. So, um, my website is, um, white light digital marketing is wldm. io. Um, and like we offer services, like I, I, I, we’ve doing a whole bunch of other really cool, like really, really cool stuff with machine learning and link building, all this stuff.
And, but it’s all unpublished because it’s all secrets girl stuff, because we’d like to keep it for our, um, clients. Um, but essentially, if they want to reach out to me, it’s brie, B R I E, at wldm. io. And that’s, that’s, that’s simply me. Brie, thank you so much for coming on. I appreciate it. Yeah, thank you, Jared.
I appreciate that. Thank you. It was a great chat.