Skip to main content

Digital Marketing

What Google Says vs. What SEOs Believe: Eric Enge and Rand Fishkin Analyze Survey Results

What Google Says vs. What SEOs Believe: Eric Enge and Rand Fishkin

A SparkToro survey of more than 1,500 digital marketers revealed a wide range of opinions and views on Google ranking factors and whether or not Google’s statements about them are true or provably false.

On November 7, 2019, Eric Enge (General Manager, Perficient Digital) and Rand Fishkin (Founder, SparkToro) presented a webinar, What Google Says vs. What SEOs Believe. They discussed the survey’s findings and provided insights on areas such as content accuracy, AMP, page speed, and more.

Find out what Eric and Rand revealed by checking out the webinar recording below or reading the transcript.



Eric: Hello, everybody. I’m Eric Enge. I am a general manager in the Digital Marketing Department at Perficient Digital. And with me is Rand Fishkin. Say, “Hi,” Rand.

Rand: Howdy, gang. Good to be here.

Eric: great to have you back, Rand. I think everybody here knows you that’s watching this. I think they do, but for those who don’t, Rand is originally the founder of Moz. And in my habitual embarrassment, I’ll refer to him again as a legend in the industry.

Rand: That’s very kind.

Eric: And it’s awesome to be able to do this together with you again, Rand.

Rand: Eric, I love joining you for this. These are always super fun, brings me back to my days back when I was at Moz. Of course, now I have this new company, SparkToro. And SparkToro has been doing some interesting research which you and I have gone through together and which we’re going to be presenting on today. So I’m thrilled to get started.

Eric: Yes. I’m thrilled to get started, too, and I just want to mention briefly that this is just a sample of some of the awesome research that SparkToro has done. There’s a lot of other really awesome stuff, too. And I’ll give you the free plug that you didn’t ask for. You need to check them out. Very awesome things that you’re doing there at SparkToro, Rand.

Rand: Thank you. Actually, I have some new data coming perhaps later this week or next week too with the Q3 numbers on Google’s click-through rates and some analysis of the Google Jobs widget after they got more aggressive with that, so, yup, stay tuned.

Eric: Awesome. Today, we’re going to start by talking about this latest ranking factor survey that you did. But why don’t I let you talk about that and we can push the first slide live, and you can introduce the whole thing to the audience?

Rand: Great. basic story here is that in August of this year, so just a couple of months ago, you know, Casey and myself basically launched a survey of Google ranking factors. And this is something that for those of you who have been in the search and SEO world for a while, web marketing world for a while, you might recall that when I was back at Moz, I started doing these all the way back in 2005, and every 2 years, I would do them again. So, 2007, there was an addition, 2009, 2011, 2013, 2015, and then there was nothing in 2017. I was a little disappointed. I was kind of like, “Come on, Moz, let’s do another one,” but it didn’t end up happening.

And so when I left Moz and started SparkToro, I decided, “Hey, if Moz isn’t going to do it, I’ll do it myself.” And so that is what happened here, essentially, collected primarily through email, through LinkedIn, and through Twitter shares. These 1,584 professionals in the search marketing space contributed to this survey. They are global respondents, although 70%, 80% are in the U.S., UK, and Canada, so mostly English language-speaking folks. And they were surveyed on two variants. One was what factors they believed Google was using and weighting and how heavily, and then also, which we’ll discuss a little later in this webinar, their opinions on Google’s representatives’ public statements about what was and wasn’t used by Google’s systems.

So there’s two things we’re looking at here and comparing them is a fascinating study in the opinions of the SEO world. That is essentially, what do you believe Google is using to rank and what do you think of what Google says about what they’re using to rank? So, yes, here you go, the Google Public Statements Opinion Survey, which same survey, published at different times. You can find the full results on SparkToro, but we’re going to be deep-diving into some areas and cutting the data in some interesting ways, so let’s get into that.

Eric: And that sounds good. In fact, we’re going to go through 10 of those things that we talked about. The first one was really… And now I gotta put on my glasses to make sure I can read the question appropriately. In any case, we asked, or you asked, rather, how much weight does web page load speed receive in organic ranking? So let’s actually look at what those results were. And the reason why I wanted to highlight this question, Rand, is it looked to be…look, we’ve got a 6.8 average rating there, which is a pretty high rating. It says that people…

Rand: Very high.

Eric: Yes. So people really believe that web page load speed is a significant factor in ranking. If that’s the case, why do they do so little about it?

Rand: It’s interesting. So your perspective is why do they do so little about it. My perspective is, “Boy, I see a lot of slow-loading sites do real well in search results.”

Eric: I agree.

Rand: Right? And Google’s public statements around this have not matched this idea that it is hugely important. I have seen, you know, Googlers say, like, “Oh, yeah, you know, we’re introducing it as a light ranking factor. It’s a small ranking factor. Yeah, we use it, we care about it. Right, you should try and make your sites and pages load fast.” But I was shocked by this result. I think I personally rated it a two or a three.

Eric: I think I was probably in the same place in terms of rating because I’ve always bought into the idea that it was a ranking factor only if you were abysmally slow, which is what Google has said about it. And the simple rationale for that is if you are by far the most relevant content for something, that still counts more than anything else like a factor like page speed. So if I want something from a particular…take a navigational search, for example. You know, obviously, we would have to show the relevant search for that. But the relevance matters so much more than everything else and we just need to get used to that.

Rand: I also think a lot of it is indirect, right? So web page load speed is something that predicts that people will come back to again, that they’ll link to more, that they’re more likely to enjoy their user experience, that they have a positive experience, they’re more likely to share you. So I think all of these things are interconnected, and this is one of the…you know, gets to the fundamental point of studying what Google actually does and studying what works to rank webpages in Google are two different practices, and I think, sometimes, we lose sight of that in the search world.

Eric: Yes, indeed. So that’s a good start to our session, but why don’t we take a look at the next question and see what that one was about? So this was another interesting one to me, and that’s about keyword use in the URL and, you know, how much weight that Google places on that. So let’s take a quick look at what the answers were for that. So that came in in around 5.2, which was an interesting result, too. I mean, that’s kind of middle of the road but it means that there’s a number of people that consider it to be a strong or even very strong ranking factor.

Rand: This one is the most bell-curvy of the ranking factor. So you essentially have a bunch of people who are clustered in the middle, right, four, five, six. And then you have kind of these, you know, outliers on the edge, people who are sure it’s nothing at all, 0 meaning not used and 10 meaning very heavily used all the time. Again, this is one of the ones where I personally was surprised. I expected to see it much lower in the rankings, and I think that there’s a holdover or a hangover belief set from kind of a prior era in the SEO world. And I think we see that we don’t experience it. Those of us on the cutting edge of SEO don’t experience how much information from 2005 to 2015 is still floating around in the consciousness of web marketers. And failing to recognize that, failing to understand that, you know, it’s our job to sort of educate the field past that era is critically important. And I think you were using the URLs as one of those ones where… Are we getting some background noise from Cindy and the folks? Maybe we should mute them.

Eric: I agree with you. In fact, one of the pieces of advice that I like to give, Rand, are when people ask me about during, you know, changing their site around and changing the URLs on their site to get keywords in the URL. I think don’t do it. The downside of changing all of the URLs, you know, outweighs any tangible…what I perceive to be maybe tiny benefit.

Rand: Eric, one counterpoint for you, what do you think about the way that Google is displaying the URLs now on mobile and in desktop results where, essentially, they appear to show kind of these categories and subcategories, and it looks like those present sort of a nudge, right, to suggest, like, “Hey, maybe you should click this result instead of this other one?” Do you think that’s still existing, that biases user behavior?

Eric: So I do think that the presence of those would bias user behavior somewhat. And, obviously, Google is testing it, so we’ll have to see, you know, how that goes for them in the long run. But I noted that the other day, too. It’s just…

Rand: I have had this… I don’t want to say it’s a conspiracy theory, right, but I have this theory. Google says, like, “Oh, the way we try to present results is so that your, you know, click-through rate…so that we maximize the click-through rate of websites,” right, that people find the answers they want, they click on them. And I kind of don’t believe that because the data shows that zero-click searches have risen so dramatically, right? And Google has basically killed off so many clicks by presenting information just right in the search so that people don’t need to click. And so I wonder to what extent the algorithms are designed to make the visualization and the data presented provide searchers with information they need without having to click versus optimizing to actually get clicks. You know, on the ad side, I get that Google has an incentive, but on the organic side, they kind of have a reverse incentive, and so I’m skeptical.

Eric: I hear you. I’m just also mindful that Google has to be very careful about how they manage the satisfaction of their users. And I suspect that they’re balancing what they keep on-site with when they send people off-site.

Rand: I think that’s probably true.

Eric: Should we look at the next question? I think you’re going to drive this one.

Rand: I like this one, too. All right. So the question here was, in your opinion, how much weight does anchor text of links pointing to the specific, ranking page impact the Google results? So this is a classic ranking factor, right? I point a bunch of anchor text with my keyword at the page I want to rank. Am I going to be able to boost it? And, you know, for 20 years that you and I were in this industry, right, ’97 to 2017, this was the ranking factor, right? You could basically just count on the fact that if you got 5, 10 more links with exact match anchor text from good sites pointing to yours, you were going to rank better. And you could test this, you know, a million ways from Sunday and see it just work every time. In the last few years, I have heard that SEOs are generally a little more divided on this. And if we look at the… Sorry, can we pull back up the distribution there? Yeah, if we look at that distribution, right, you can see that there’s still…it is a strong average rating but it’s not even as high as webpage load speed. And, you know, if you took the ranking factor survey in 2005 up to 2017, this was, if not number one, number two or three in the ranking factors. Now, I think it’s number seven or eight. So it is all in dramatically in the eyes of professional SEOs.

Eric: It absolutely has. And, by the way, I’m in the camp of those who believe that it matters a lot less than it used to. We have seen people do very well with strong link profiles that weren’t anchor text-rich, and we’ve seen that many, many times. But I’m not saying it’s not a ranking factor at all and I think it depends on how they look at it.

Rand: Obviously, for these 20 years that it was super important, Google, you know, obviously, relied on it. They needed it, they used it heavily. What do you think they have replaced this with? Like, what is the data that they’re getting now that they’re able to say, “Hey, we don’t need to rely on anchor text exact match as much anymore because we have?”

Eric: If so, I could tell you what I hope they replaced it with. Not necessarily what they have replaced it with, but maybe the relevance and the context of the link a little more broadly. That would be a nice thought, at least, that…

Rand: I mean, I generally agree. I think that they’ve gotten way smarter so that if you go to…you know, if you’re trying to rank a page about, you know, Italian cooking and you go to a bunch of pharmaceutical websites and put Italian cooking in the anchor text, Google just is not going to care about that like they used to because they can see that the page and the website has nothing to do with it and that there’s no relevance, no connection there.

Eric: Absolutely. We should look at the next question.

Rand: In your opinion, how much weight does accuracy of content… This is an interesting one because we’re going to actually come back to this in a future one, but we’ll talk here just about the opinions of SEOs. In your opinion, how much weight does accuracy of content receive in Google’s ranking systems? Let’s take a look at the distribution. Oh, right, before we do that, a public statement from Danny Sullivan who many of you might know, he used to run Search Engine Land and Third Door Media and then was hired by Google as their sort of internal search liaison, which I’m very sad about because I loved his voice as a Google critic but he certainly provides some interesting information from inside of Google as well. So he says, “Machines cannot tell the accuracy of content. Instead, our systems rely on signals we find that align with relevancy of topic and authority.” So this first sentence, Eric, I just want you to rate the truthfulness for me of the first sentence from Danny here, “Machines cannot tell the accuracy of content.”

Eric: Let me try a sample question for you. You could tell me whether a machine could tell the answer. The capital of Washington State is Seattle.

Rand: That’s false, and I believe every machine could tell you that.

Eric: Clearly, there are some lines where they can tell the accuracy and others where they can’t, right?

Rand: I mean, I find this such a strange statement from Danny, right? Because I think that, historically, he’s always been very concerned with perception and sort of, you know, the accuracy of his own statements, and so I found this one very odd. However, let’s take a look at the distribution. All right. Well, SEOs, obviously, unlike Danny Sullivan, believe that Google can not only rate the accuracy of content but they believe it is extremely important. This was one of our top five ranking factors in the 2019 survey.

Eric: All right. So that’s great. Let’s take a look at the next question. So for this one, what we really want to find out is in your opinion, how much weight does the total amount of the content on the page receive in Google’s organic ranking systems? This is a fascinating question, too. So it is kind of a middle-ish of the road score, but that means there’s a lot of people that believe that the volume of content actually matters. And I find this one fascinating, too.

Rand: I think this is another example of correlation-causation problem, right, where it is absolutely the case that you can see that more content correlates with higher rankings, right, with more page one on top of page one rankings. No doubt about it. Every correlation study that’s ever been done in Google’s rankings in the last five years have shown more content. But I think what’s hard to know is, is that cause or is that merely a correlation? And I tend to think it’s more correlation. I don’t think Google actually gives two poops about how much content is on the page, and they only care that it solves the searcher’s problem. And for in-depth queries, many in-depth queries, there is quite a bit of content required to solve the searcher’s problem.

Eric: Right. But I believe that what’s maybe at the heart of this, too, is if we think about what Google needs to do and you go back to a good part of what makes them successful is the satisfaction level they maintain with their users. In-depth answers to questions are often what users are looking for even though there’s a strong culture out there that says less is more, and soundbites, and snackable content, which I think is all BS, personally. I mean, there’s a time and a place for it, don’t get me wrong, but…

Rand: It’s called Twitter.

Eric: That’s where you get the snackable content. So, you know, I think there’s an underlying element of truth to this, but you can’t turn that into you need at least 250 words on every page.

Rand: This one was a little disappointing for me. To be totally honest, I would have loved to see SEOs say, “No, it’s obviously not a ranking factor on its own even if it’s correlated.” And maybe as people get more sophisticated around this subject, we’ll see that in the years ahead. So our next question.

Eric: All right. Is that one for me or for you? Oh, it’s me again. So this is another fascinating one. In your opinion, how much weight does the presence of external links in the page’s content receive in the Google ranking factor? It’s like, okay, linking out is a positive for ranking, is what this question is about. Four-point-seven which, you know, slightly less than average, obviously.

Rand: That is quite low. I think that’s in our bottom five or six.

Eric: Right. But on the other hand, I could tell you if I look at this distribution, this is not equal to the distribution of people who actively link out from their content.

Rand: I think what’s interesting, right, is I hope everyone links out from their content because I believe that it is similar to load speed, similar to amount of content on a page. It has a strong correlation with ranking well because when you link out, not only does that… I do believe it is a ranking factor although a small one, but I also think it shows the rest of the web that you are a good participant in the ecosystem, and that tends to come back to you, right? Just as networking and building up friendships and relationships in your industry can have a positive impact on your career and almost always does, so too does building up relationships with the rest of the web, especially in your own ecosystem and showing that through external linking.

Eric: I agree completely that it’s a really smart thing to do.

Rand: And I think it helps searchers solve their problems, to your point earlier.

Eric: If you provide the…

Rand: The satisfaction is right there.

Eric: Right, absolutely. All right. Let’s take a look at the next one. These are kind of paired up, which is what I find interesting, the next two questions. But in your opinion, how much weight does the use of Google AMP web component framework receive in Google’s organic ranking systems? So let’s go ahead and look at the answer, and this is another one of those industry contradiction…

Rand: We’ve got even distribution, right? Basically, you could just get very, very even across the board. This is one where there’s incredible disagreement.

Eric: It’s true. But you’ve heard me say this on two other questions so far. If this is really the distribution of people who believe that AMP is important to ranking, then how come half the websites on the web aren’t in AMP?

Rand: I reject it for philosophical reasons.

Eric: Well, okay.

Rand: But, yes, I agree with you. I think that…well, let’s move on to the next question, and I think that’ll actually answer…

Eric: They’re tied together so go for it. Let’s see the next question.

Rand: All right. So, this one, we’re moving into the how do you feel about Google’s public statements. Here’s the public statement from John Mueller who works at Google I think as a search or webmaster…

Eric: Trends analyst.

Rand: Analyst. And John says, “AMP isn’t a ranking factor; if you decide to disable it, make sure to redirect appropriately.” I agree with the last part. I don’t know about the first part. But how truthful do you rate this statement? And here’s the distribution. So, first off, it’s a zero to four scale, which could be a little awkward for some folks. Zero is, “I believe this is provably false, I can prove that AMP is a ranking factor,” and four is, “This is 100% transparent and correct.” And then two is…I used the acronym TCBM, but all of the raters saw what it actually said which is, “Technically correct but misleading.” So that is where most of the answers came in here, technically correct but misleading is what most SEOs believe. And then you can see there’s almost exactly the same number who say it’s provably false versus those who say it’s 100% correct.

Eric: No, it’s fascinating to see this distribution here, too. And I think part of what people are pointing at is, of course, if you’re a new site and you get in the top news or top stories carousel, yeah, for those sites, it’s clearly a ranking of factor. But, you know…

Rand: Which really bothers me, right? It really bothers me that John Mueller, who should be a responsible person, right, who presents accurate and transparent information, I think, you know, that’s his job, that’s how he’s sort of judged, that he would exclude that critical information when he’s making these statements. And he didn’t follow up, he didn’t correct it. When he was confronted about it, right, all of the replies say, like, “What about Google News?” And there’s no reply from him. It’s frustrating, right? I think it causes less trust to be in Google’s reps. I wish that wasn’t the case. That frustrates me a lot.

Let’s take a look at one of the Google News results here. There you go. I see, let’s see, one, two, three, four AMP results, one, two, three, four AMP results, one, two, three, four AMP results. Basically, there’s nothing but AMP. Because you cannot appear in these types of results, literally cannot in these types of results unless you were AMP.

Eric: Now, just to…by the way, I agree that it would have been great if John clarified it. But I will say is outside of the world of news sites and getting into these kinds of results, I don’t see any real evidence of AMP being a ranking factor for what I’ll call “the rest of us.” But, you know, these contextual nuances are really important in communication, you know? There is no…

Rand: I think it’s really important to maintain accurate communication. Let’s take a look at the next one. How truthful do you rate this statement? Again, this is from the same person at Google, John Mueller. “There’s no such thing as LSI keywords. Anyone who’s telling you otherwise is mistaken, sorry.” Here’s the results. Most people, again, believe technically correct but misleading. This one I have a little bit of forgiveness for. I think that when John is speaking about this in the context of does Google use latent semantic indexing, the old-school technique from, like, the late 1990s for their particular, you know, analysis of words at Google, no, they don’t. Is there such a thing as LSI keywords? Yes. Yes, there is, right? If I Google, “What’s an LSI keyword,” the results won’t all say, “There’s no such thing as that.” So a little bit, okay, it’s technically correct but misleading, but we can forgive John, I think, in this case because in my opinion, at least, I don’t know about yours, but I don’t think Google uses this at all.

Eric: So maybe TCBM isn’t right. Maybe it’s technically incorrect but underlying truthful.

Rand: I should have surveyed for that, too. What I find interesting here is a lot of people…more people have said this was provably false than 100% correct, and I believe that is because many folks have seen tools and case studies and have tried it themselves where they basically gone and found related words and phrases, maybe even using the LSI protocol or system. They’ve included those words and phrases on their pages and they’ve seen rankings boosts, and that is…yeah, that’s [inaudible 00:27:55] stuff, right? So I agree that, sometimes, even when you’re not using the system Google uses, you can still get positive results out of it. I think there’s just some more sophistication that has to go on there.

Eric: All right. Shall we take a look at the next one?

Rand: How truthful do you rate…? Ooh, I love this one.

Eric: This is our final one. This is a great one.

Rand: Oh, I love it. All right. How truthful do you rate Google’s statement? “We don’t have anything like a website authority score.” So let’s look at the evidence here.

Eric: Survey said… Oh, wait, we have to see…

Rand: Before we look at that, I just want to point out, right, so this statement was made officially by Google. I think it was made by Gary Illyes?

Eric: Illyes.

Rand: Illyes? And then just a few months later, “The Guardian” wrote an article where they cited Pandu Nayak who’s been at Google for 14 years who works as a search engineer. And so he says, “In order to address this, we have developed algorithms that recognize a bad event is taking place and we should increase our notions of authority, increase the weight of authority in our ranking so that we surface high quality content rather than misinformation in this critical time here.” Which sounds like this search engineer who works at Google is literally telling “The Guardian’s” reporter, “We have a concept of website authority and we are increasing its weight in our rankings.” So maybe SEOs can be forgiven for not believing the Google statement at all, right?

Eric: Provably false.

Rand: This is our least believed statement where most SEOs looked at that and said, “That is a pile of trash and we don’t believe you one bit.”

Eric: Interestingly enough, there was another statement when they were talking about visual search where they made a reference to the authority of the webpage having to do with whether an image would be picked in a visual search context. So the statements have come out that sort of contradict this other statement. So this is one of those things where I do think a lot of the time…and this is where your true but misleading context is a really good one that you put in the survey is that, okay, maybe since we all associated this, like, purely with link score, page rank, okay, maybe that association isn’t the way we all think of it.

Rand: Or maybe it’s not a single website authority score, right? Maybe there are multiple algorithmic elements that combine to suggest something is more authoritative than something else, right? And that could make it such that the statement is technically accurate. What I don’t love about this statement is it doesn’t say, “We don’t have a website authority score,” it says, “We don’t have anything like a website authority score,” which suggests there is no concept at all inside Google of rating websites on their authority. And that feels very…I don’t want to say, “Misleading,” it just feels like a direct lie, which, again, I think this sort of hurts credibility and you can see SEOs kind of believing that too.

So let’s move on and wrap up here. We’re going to get to some Q&A as well. You know, my closing thoughts on this…and, Eric, I would love yours as well, but my closing thoughts on this are I think that in many cases, paying attention to Google’s statements is wise and it can have interesting value, right? Like, we as marketers can derive interesting value from them, but we can’t take them at face value and we shouldn’t take them uncritically. And I think it pays to look at how the field distributes their answers and to see how disparate web marketers are in their beliefs about what causes rankings, whether Google’s statements are true or not. I find that information fascinating. I think that suggests to both marketers and anyone who uses marketer services that there is a huge delta in our field in terms of what people think and how they think it works.

Eric: And I think that when we look at the John Mueller statement, for example, about AMP as a ranking factor, you know, there’s, I believe, this internal culture thing at Google where what people are allowed to say and not say is constrained and maybe someone makes a public statement and some hornets’ nest gets stirred up around that. And I do think there are times when they get told, you know, “Let it die down,” rather than actively addressing it. And I do think Google would be better served with, you know, just clarifying these kinds of things, so I agree with you there. But the other thing…

Rand: And I also think…I worry a little bit… You know, if I were a PR person giving advice or if I were a manager, right, on the Google representative team, the advice I would be giving to my team, my employees would be, “Hey, it’s okay to say, ‘I’m sorry but we don’t reveal that information.'” And that’s a statement you almost never see from Google anymore, right, since the Matt Cutts era when Matt would say that frequently, like, “Well, I know the answer but I can’t tell you.” And I think we could return to that. I think it would be absolutely fine and fair to say, “Hey, that’s a great question. It’s not something that I’m authorized to give out.”

Eric: The other thing I want people to think about a little bit too, while I agree with you saying not to place too much weight or be over-literal on how you interpret what Google says, there’s times where understanding that they are, in many cases, very literal about how they’re applying the terms is really important. And understanding that landscape can be critical in how you get the most value from Google statements. So a classic example is when Google said that we don’t use bounce rate as a ranking factor. And I remember Matt Cutts going on this long rant that [inaudible 00:34:26] makes advance about bounce rate and he concluded it… It was actually a question from the audience, “Do you use bounce rate as a ranking factor?” He did a long rant and then he finished. So, in conclusion, we don’t use Google Analytics bounce rate as a ranking factor.

Rand: Right.

Eric: Okay. That wasn’t the question.

Rand: That’s successively technically minded answer.

Eric: Right. And so you just really need to understand or try to read through the lines to what the real context of where they’re coming from is. It’s not easy to do but definitely worth the effort. But let’s go to the first question.

Rand: We’ve got some great questions here.

Eric: The first question is “How do you explain to clients the massive shake-ups from the last Google update despite doing everything ‘correctly’ the white hat way?”

Rand: I think that this happens…I don’t actually know which shake-up this person is referring to. You know, there was a big shake-up in July, there was a big shake-up in May, there was a big shake-up in September and another one in late October. My guess maybe is that this question is about the most recent one. But in any of these cases, the good news is, I don’t know if you’ve seen these charts, right, but a lot of the times, Google will make a big change. Some people will lose rankings, some people will gain rankings, and then over time, that will balance back out again, right? So in the next shake-up or in the correction to a shake-up, oftentimes, someone will benefit.

The other thing I’d say is that doing things the correct white hat way is not a winning tactic anymore. It is just table stakes. In order to win, you have to be doing things vastly better than anyone in your field, right? So if you tell me, “Hey, I haven’t done any bad link building, I haven’t done any sketchy content stuff, I’ve done no sketchy on-site stuff. Why am I not ranking better?” I would say, “Those things do not help you rank better than your competition, they only make sure that you don’t get penalized.” And these algorithmic shake-ups are not about penalization as much anymore. Certainly, not large-scale since, you know, the days of Penguin and that sort of thing. Mostly what we’re seeing now is shake-ups based on who’s solving the searcher’s problem better than anyone else.

Eric: I couldn’t agree more, and this aligns so well with what advice I give people on things that I say in presentations. I mean, your job is to, just to give it a simple phrase, be elite, be the best at something. It doesn’t have to be your entire market space, by the way. So if you’re in a market where there’s three, you know, giant corporations and you’re a mid-tier size player and you’re trying to gain some market share, take one slice of your market space and go deeper, and further, and more thorough on how you address that piece and be the best at that piece of it. Short of doing this, as Rand says, the white hat SEO thing is just table stakes. It isn’t going to get you there. All right. So the second question.

Rand: Our first question here was, “Are these ranking factors and inputs universal or does Google use some only in certain cases?” And this is an amazing question because one of the things that I asked in the survey was, “Do you believe Google has a universal set of inputs or that they have some subset where they apply different factors differently or do you think that they apply them differently to every different query?” And SEOs gave their opinion on that and, by and large, most folks believe that there is some bifurcation, right, that, essentially, you have variants in how Google weights the algorithmic inputs based on the type of search. And you can see the most extreme example that we showed today was the AMP results, where in news, AMP is not only a ranking factor, it’s a requirement. It is the ranking factor, right? Either you have it and you’re allowed in or you don’t have it and you can’t get in at all. And that is true in, I think, many spaces not as extreme as AMP in news but true in many cases. So great question there.

Eric: I believe there’s a pretty significant variance in types of results and kinds of behavior. You could actually see this in a very simple way. You could use a set of tactics that works really well for one business and you turn around and do exactly the same thing for another business into another market and you get pretty different impact. So there’s definitely, you know, a different nuance there. So the third question, Rand, was, “Does the balance of no-follow and follow links in your external links matter?”

Rand: If you ask me, personally, the answer is mostly no with one exception which is if all of your external links are no-follow. If you blanket no-follow everything, I think that that is going to have a negative impact on you. Whether it’s a ranking factor, I can’t say for certain, but I have seen it being negatively correlated with performance and rankings, and that could very well be from that ecosystem participation. It could be that Google looks at it and cares about it. The only site that historically has performed really, really well with blanket no-follows was Wikipedia, and they’ve actually done significantly worse in the last 4 or 5 years in Google’s rankings than they did for the first 15 years of Google’s life.

Eric: And I think, you know, most users aren’t going to notice no-follow versus follow for sure, so they won’t be aware.

Rand: But the users who link, which matter a lot, they will notice.

Eric: They will notice, I agree with you. But from the point of view of the deeper ecosystem, and was I willing to have someone leave my site and go to another site, which is kind of a big deal, when you think about the value of an external link from that perspective, that piece is cleanly passed through. So our next question, “The most surprising belief or opinion,” what did you think?

Rand: Okay. So if you go to the actual chart, you can Google, you know, SparkToro ranking factors or Google ranking factors 2019, you’ll find it. One of the things that I found fascinating was the relative disagreement, right, so the disagreement levels. There were some things where people very much agreed, so the standard deviation rate was quite low. For example, almost all SEOs believe that relevance of overall page content, that’s super important, and so the standard deviation on that was super low. Keyword in the domain name is one of the ones that I found maybe most, I don’t know, surprising, but that was the most disagreed upon ranking factor. There is still a strong cogent of search marketers who think that having the keyword in your domain name, not your URL but your domain name, is a big important ranking factor and then a bunch who don’t. And it’s kind of…you know, whereas bell curve is their standard distribution, this is like two towers on either side, you got a bunch of zeros and a bunch of fours.

Eric: No, I thought that was a fascinating one, too. And, I mean, there were so many, though, and we actually had some of them in here where the distribution was really pretty even, you know, there’s a sort of flattish shape. So just like you said earlier, it just reminds us just how much disagreement there is out there, and, frankly… And I’m not saying that was the case in these particular questions, there’s still so much disinformation about SEO out there and varying opinions with things that aren’t really…there really isn’t any real data to support them. So just know that you have to be careful out there when you’re… Pick the people who you’re reading for SEO information and advice carefully.

Rand: The only other one I was going to bring up is this stuck-in-the-middle ranking factor that I thought should’ve been near the top which is freshness and recency of publication, which, personally, I’ve seen in almost every space have a lot of positive impact. And so I was surprised to see that one stuck in the middle. I think that might be an underrated factor, and potentially, therefore, a competitive advantage for those of you seeking one of those.

Eric: Yes, indeed. So, anyway, we’re at the end of our time, Rand. Awesome as always.
Rand: Eric, I love joining you for these things. It’s so much fun. Folks out there, if you have questions for us, you can shoot them over to me, @randfish on Twitter. And, Eric, you’re @ericenge now on Twitter?

Eric: No, @stonetemple still.

Rand: @stonetemple on Twitter. Okay, great. Excellent.

Eric: All right. So, thank you, everybody. I hope you enjoyed the show, and, as Rand says, feel free to shoot questions over. We’re always happy to take them on.

Rand: Awesome.

Eric: All right. Bye now.

About Rand Fishkin:
Rand FishkinRand Fishkin is the founder of SparkToro and was previously co-founder of Moz and He’s dedicated his professional life to helping people do better marketing through the Whiteboard Friday video series, his blog, and his book, Lost and Founder: A Painfully Honest Field Guide to the Startup World. If you feed him great pasta or great whisky, he’ll give you the cheat code to rank #1 on Google.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Eric Enge

Eric Enge is part of the Digital Marketing practice at Perficient. He designs studies and produces industry-related research to help prove, debunk, or evolve assumptions about digital marketing practices and their value. Eric is a writer, blogger, researcher, teacher, and keynote speaker and panelist at major industry conferences. Partnering with several other experts, Eric served as the lead author of The Art of SEO.

More from this Author

Follow Us