What Cyrus Shepard Found After Analyzing 23M Internal Links

Do you want to know the optimal number of internal links you should build to new pages?

Or how about which anchor texts will get you the best results?

In this episode of the Niche Pursuits podcast, long-time SEO expert Cyrus Shepard shares some very interesting answers to these questions you won’t want to miss.

His findings come from an in-depth study he did on the impact of internal linking on a website’s ranking and traffic.

He analyzed 1800 websites and 23 million internal links and found a positive correlation between internal links and Google traffic. But there’s a catch. After a certain point, the relationship between internal links and traffic diminishes.

So, how can you make the most of your internal links? Cyrus emphasizes the importance of controlling both popularity and relevance in internal linking. He shares insights on Page Rank Sculpting and advises starting with a plan and adding links as soon as possible to avoid playing catch-up later. He also believes updating old content and internal linking should be a continuous part of your site management process.

Cyrus then shares tips on where to place internal links and the importance of varied anchor text. Plus, he shares his thoughts on keyword cannibalization and an example of how linking multiple pages together using keyword-rich cannibalization keywords improved the ranking of all pages.

But internal linking isn’t the only factor at play. Redirecting 404 pages with a 301 redirect to a highly relevant page is important. Plus, he recommends automation tools like Link Whisper to help find internal linking opportunities.

In short, internal linking is an essential aspect of SEO, and this episode is packed with actionable tips to help improve your website’s ranking and traffic. So, tune in now to learn more and start testing out these strategies on your own sites!

Topics Cyrus Shepard Covers

  • Cyrus’ experience at Moz
  • Running experiments
  • Quick SEO wins
  • Link popularity vs relevance
  • Importance of anchor text
  • Link position
  • Dangers of duplicate anchor text
  • Naked URLs
  • Avoiding cannibalization
  • Page sculpting tips
  • First link priority
  • The magic number for internal links
  • Inbound linking tips
  • Planning internal links with content creation
  • Related links position
  • Using images for internal links
  • 404s, 301s, etc.
  • In-text links

Links & Resources

This Episode is Sponsored by Search Intelligence & Nichesites.com

Watch The Interview

read the transcription

Jared: All right. Welcome back to the Niche Pursuits podcast. My name is Jared Bauman. Today we’re joined by Cyrus Shepard. Cyrus, welcome on board. 

Cyrus: Hey, Jared, thanks for having me. Excited to be talking with you today. 

Jared: I am really excited to have you on board. I have been following along at zippy.com with a lot of your different studies and deep dives into different parts of SEO over the years.

I’m so glad to finally be able to connect. And today we’re gonna kind of zero in on one specific topic. Before we do though, maybe give us some backstory on who you are and how you got involved in this industry and, you know, kind of catch the readers up and the listeners up to who you are. 

Cyrus: Yeah, sure.

I’m Cyrus. I studied film in college, which is a perfect segue to marketing because I failed as a Hollywood screenwriter. But I got my start in SEO the way a lot of people did. I had, I had websites, I was trying to market them. Didn’t wanna spend a lot of money on AdWords at the time. And I ended up joining this company called SEO Moz in 2011.

It’s known as Moz today. Back then, you know, it was, it was like the biggest s e o company on the planet and we just had a lot of fun. It felt like every day going into work, we were inventing the internet. And at Moz I did a lot of large scale experiments, or I ran a lot of large scale experiments, which, you know, back in the day, we tested Google’s algorithm, I think a lot more than we do today.

Maybe because it was simpler back then or Google. That people were just trying to figure things out and I, I really missed that. So today, in my career I tried to run experiments whenever I can. Publish them whenever I can. And I think it, I think it’s the most fun thing that I do. First, 

Jared: fun fact, I didn’t know it was originally called SEO Moz.

I always knew it just as Moz. 

Cyrus: Yeah. So yeah, Moz founded in 2004 by Rand Fishkin and his, his mother. Yeah, it was, it was, the idea was Moz was sort of Mozilla. It was kind of the open movement, the open internet movement, and they were, you know, trying to be a part of that. So, yeah, I always say SEO companies are terrible at choosing names.

Yeah. Well, 

Jared: we’re not too creative, are we? You can, I actually have a photography background, so that’s really interesting. I don’t meet many people from the creative space that. In seo, it’s kind of a right brain, left brain thing, but yeah. What was so what was your role at Moz and what did you end up doing at Moz for the time you were 

Cyrus: there?

So I started off in customer service, customer support. I also unloaded the dishwasher every morning at 6:00 AM I was the only one in the office answering phone calls. Hello, hello you know, helping people with their, their porn sites, whatever, whatever it was that they had signed up to seo oz.

And then I transitioned over to the marketing and doing actual seo and I led SEO in content for several years. I had three, actually three different stints at MAs. I always came back to the job as the job. I could never quit left for the final time when they were acquired by eye contact marketing group a year or two back.

And but still lots of love for the old 

Jared: company. Yeah. Yeah. Well you, you, you’re now at Zippy, which is your own company. It’s an SEO consulting company. You guys also make soft. You, you mentioned you transitioned over the course of the last good many years. I guess you said you had many assets from ma, but you know I guess to set the stage for today, the, the study that you did in particular that caught my eye and I think is gonna be a perfect Almost like case study for us to talk through today is your study on internal links and I I will say it’s probably not a very popularized study subject, right?

Like we get a lot of studies on back links, we get a lot of studies on intricacies with back links and keywords and rankings and on page stuff. Obviously internal linking could be considered linking, could be considered on page, but it’s not studied that much. What gave you the impetus to study internal links and and, and produce this?

Yeah, great 

Cyrus: question. So whenever I talk to agency owners and we’ve, we’ve sort of chatted about this before a little bit. When, whenever p agencies need to move the needle quickly for clients cuz. Clients don’t wanna wait six months to see results. They need to move the levers that they can control very quickly.

And, and generally those are you know, metadata, title tags, meta descriptions, things that show up in search pretty quickly. The other lever that they can pull really quickly is internal links because you can link to yourself much quicker than you can asking someone else for a link, and you should still do those things.

Improving your internal linking is one of the fastest, most effective ways you can do for a quick win. So that’s, those are the first two studies that we ran were title tags and internal linking. Ah, I saw the 

Jared: title tag one. I was gonna. If we had time. I have that down, but, but today we’re talking internal links.

I mean, we do talk internal links a decent amount around here. Obviously Spencer, the, the founder owns Link Whisper, which is an internal linking plugin. And I helped test that when it was first started. And it’s such a internal linking can be so frustrating, especially like you said, for agencies or for people who are trying to.

You know, build links internally at scale. So anyways, before I bury the lead, like let’s dive into it. Let’s talk about this. You analyzed, I think you said somewhere near 2000 websites 23 million internal links. Like why don’t you set the stage for the, for the, for study and we’ll get into some of the findings.

Cyrus: Yeah, and I wanna circle back to Link Whisper and other tools, which I recommend all the time and why I think we should use them. But first, let’s set the stage. So yeah, so you often see studies on linking, external linking and I think one of the reasons is those studies are. Much easier to run.

Mm-hmm. Having, having sort of run a few myself, you, you, you can get some, you know, data from a refs or some rush and you know, compare it against rankings, things like that. But we wanted to dive a little bit deeper. So we had 1800 websites 23 million internal links, and we had access to Google Search Console, which these folks that opted into, so we could look at anonymized Google search console data so we could actually correlate.

Internal link, our internal link metrics with actual Google traffic and we can see what’s working and what’s not working. And, and that surprisingly, we found some positive relationships between internal links and, and Google traffic. That’s, that’s the very first thing we found. 

Jared: So more internal links, more traffic.

I, I suppose that’s obvious. Is there, are there any. Great points, thresholds, or anything beyond? Just more is better. Yeah. 

Cyrus: Well, first of all, we’re just happy to see the positive correlation that Yes, fair enough, more internal links actually works. It actually increases traffic. If you have a page with no internal links, it’s, it’s, it’s usually not gonna get a lot of traffic.

But it was, you didn’t if 

Jared: you didn’t say that. 

Cyrus: Yeah. So it, but the funny thing we found was the relationship only worked to a point we saw. This graph that went up and to the right. And then after about, you know, 20, 30 links, it, it just sort of like zigzagging up and down. And there we discovered the reason for that.

And it had more to do with the type of internal link that was, that people were using. So that, that was interesting. 

Jared: Let me ask you this question. It, and if this is going too deep, too fast, tell me if we’re gonna get into it a little bit later, but yeah. It makes perfect good, perfectly good sense that if a page has zero internal links pointing to it, obviously that’s bad.

It makes it highly UNC crawlable for Google, especially as it ages, it gets buried in your site’s architecture. However, why is in theory more is always better, but why, why is more better and is there a certain limit? You talked about how at some point that there’s diminishing returns, like is there a certain limit to the number that are positively effective for the, for the page.

Cyrus: Yeah, so I’m gonna get right into it here. But first I want to preface this. What we learn about links is a little bit misguided, I think in the SEO industry because we’re always talking about page rank or link Jews, blah, blah, blah, whatever term you want to use. Every, every tool has its own metrics, domain authority.

The problem is, The problem, and this is based on Google’s, you know, original page rank paper from many, many years ago. But the problem is links actually pass two, two major classes of signals. One is popularity, which is page rank or whatever we think of is, you know, Link juice, link power. But the other is relevance, which is mostly derived from anchor text.

Yep. And SEO tools. You know, Moz is certainly guilty of this A res. All the major tool providers are guilty of using the popularity metric and pushing that in their interfaces. Making it what people use, but we don’t look at the relevancy part. And that’s like 50% of the link. So when you have internal links you can control that relevancy to a certain extent.

Now there’s indication that Google doesn’t count internal links as much as external links, but you’re relying if you’re not. Internally linking, you’re relying on other people for that, for that both the popularity and the relevance. And it’s really hard to control the anchor text that other people are, are linking to you with.

So we looked at both popularity and relevance, and when we started looking at anchor text, that’s when the real interesting relationships started popping out. I’m, I’m, I’m talking too much. I’ll, I’ll let you pop in with a question if you want. No, you’re the, you’re the, 

Jared: you’re the value here. All right.

The less I talk, the better. All 

Cyrus: right. So I talked about how we saw this. We had this graph that went up and to the right. Yep. And then after a certain point, it, it petered out. What we found was those petered out links were navigation links when, when a site links in the navigation, you know, in the top navigation, you, if you have a thousand page site, you have a thousand internal links, all with the same anchor text.

And we saw a diminish. Returns on those types of links. But when we looked, then we looked at something different. How many different types of anchor text was being used? So different that means different phrases. So click here. Best camera Nikon 3 78. We looked at how many different types of anchor texts each link had when we charted.

The chart went up and to the right and it never came down. We ran the data three different times to to all the way to the right, and the, the relationship kept going up and up and up. So, To summarize this, to try to distill it down, it’s not the number, sheer number of links in our data that we found had a positive relation.

It was the number of different types of links and the variety of anchor text that made the most significant difference if a page ranked higher or not. Ugh. Hopefully, hopefully that’s easy to understand. I, the problem 

Jared: is, I have about 20 questions already, but this is great. So number one, more in general, more more internal links led to higher.

Yep. Number two, varied anchor text led to higher 

Cyrus: traffic. Yes. More importantly than just the number of links. 

Jared: Just important. So, oh, more importantly. Okay. So having a good number of internal links with varied anchor text is the proverbial holy grail of internal. 

Cyrus: Yeah. Yeah. And you think of it like this. Let’s go back to that.

You know, that navigation text again, if I. If I link to a page in my navigation, I get the same anchor text. I get a, I might have, I might have a thousand links on my site going to that page because it’s in the navigation, but the anchor text is the same every single time. Yep. So I get that, I get that popularity, but I only get, you know, one chance at a, at a relevant signal.

You know, there’s different ways of thinking about this. You could think, wow, goo I have a thousand links pointing to that page. In my navigation, I have a thousand anchor texts that say best cameras. Google must really like those thousand anchor text. If I was Google, I would just count that as one anchor text.

I would, you know, consolidate that and canonize that into, you know, one signal. And so yeah, I have a thousand. But I really only have one anchor tech, so I, I might want to vary that up to give Google more information. And we find when people do that it, it tends to have really positive 

Jared: results. You mentioned navigational.

Are there any other types of internal links that probably aren’t weighed as heavily? I’m thinking footer, I’m thinking related posts. I’m thinking sidebar. Yep. I don’t know. 

Cyrus: More. Yeah, so I always tell people years ago, years ago, I was listening to Matt Cutts who was the former head of spam at Google, and he said something, and I can’t find a reference to it anywhere on the internet.

I swear it happened. But he was like, pay attention to that first link on your page. And then he stopped and he said, pay very close attention to that first link on the page. And that’s the idea of reasonable surfer, that Google, Google Pat can pass. Wait through links that people actually click or are likely to click.

And I was just reading that patent the other day. And you know, they can, me, Google can measure the links that people click through Chrome, the, the most widely adopted browser in the world. They can see what people are clicking on. They don’t need to measure every click. They can just start, they can just record enough that they have a machine learning model and so when they see a new page, they can apply their link waiting algorithms and underst.

People are gonna click links that are high up on the page prominently displayed as opposed to, you know, links in the footer, links in the sidebar, links in, you know, top articles. So when I’m internally linking and I have something important, I put it in the first couple of paragraphs of my main body text.

If, and, you know, links to competitors, I’ll put, I’ll put down below. But I, I try to keep them high and tight kind of a military term high and. Prominent that make, make sure people are clicking on those links. And you can look in your analytics to see how many pages per visit people are visiting on your site.

And if you have those great internal links, you know, you can see those metrics start to go up sometimes because you know, people are actually clicking on them. 

Jared: In terms of, and you touched on it already in terms of having your example where you have the same anchor text in the navigational. A thousand times, right?

Yep. So you have that nice dropdown menu and it links to a, a page on your site. And because that loads on every page of your site, cuz it’s the menu, it’s gonna send a link from every page. It’s the same. It’s duplicate anchor text. Yeah. You talked about how that doesn’t probably have much weight to benefit you.

Does duplicate anchor text at any point, even outside of Navig navigational linking, have a can it actually hurt? 


Cyrus: yeah, yeah. There was definitely the idea of over optimization, especially for smaller sites that don’t have a, a diverse link profile. Now, I’m gonna take a sidetrack here for a second.

Large sites large. Retailers, things like that, they can get away quite easily with a large navigation mega, mega menu, things like that. And, and part of the reason I believe, I, I, I can’t confirm it, they already have a diverse and broad anchor text profile. If you look at sites like, you know, Walmart, look at any page on Walmart or Amazon or any, any major e etailer and look at their back link profile.

Every page has way, way high diversity of anchor texts. You as a small site and me as a small site, we don’t necessarily have that. So we might want to pair down our navigation and do more InBody linking as opposed to relying on that, on that top level navigation. That, and that gives us greater, greater control.

I lost the thread. I forget what the question was. No, 

Jared: you’re on, you’re on the right. You’re on the right track. I’m just trying to zero in on this idea of having a lot of these same anchor techs. Yes. We talked about how it has low value. Yeah. But does it have na, could it hurt you, I 

Cyrus: guess? Yeah. And again, if you’re okay, exactly.

I, I know where I was going now. And so, again, if you’re a smaller site and you don’t have this huge bling profile of external sites and you start linking to yourself with optimized anchor text that’s, you know, the same on, you know Best Nikon camera 2023. And it’s, it’s a money, money term. And you do that on every page.

Yes. That can hurt you. It looks over optimized to Google. We don’t, we don’t know exactly how Google does this and we don’t have a lot of validated proof, but we, we do understand Google can, penalize sites are being over optimized and that’s, that’s one of the key ways you can shoot yourself in the foot is using over, over-optimized anchor text every single time.

Yeah, you’re right. And we hear 

Jared: about it more on the external backlink. You know, anchor text from an outbound, or sorry, anchor text from other domains to our website is much further scrutinized than our internal linking anchor texts. But it still has an effect, is 

Cyrus: what you’re saying? Yeah, a absolutely. So the way I like to do it internally linking Is I, I don’t think there’s anything such as a bad link, but I like, I, I personally underoptimized every internal link that I can.

I make sure there’s some keywords in there but I tend to go you know, the ideal internal link is like three or four words long. Usually, you know, something clickable. I like to go. Make sure I ha add an extra word or make sure it’s a phrase that Google hasn’t seen before in my backlink profile.

So, you know if, if the keyword was best Nikon camera in 2023, I certainly wouldn’t link to it with those keywords. I’d be like, you know, Check out these cameras that we think are the best of Nikon. I know I’m giving a terrible example here. You need a little more creativity than that, but I would make sure it’s varied every time and not perfectly 

Jared: optimized.

Good. So that’s my next question is walk us through what it looks like to do anchor text properly and maybe tackle it from this standpoint. Is it more important in concept that I think of my anchor text as a way to pass relevance or as a way to help me rank for certain. 

Cyrus: That’s, that’s a, that’s a great question.

You, I don’t, I think you can do both. The way, the way I, I like to do it is I will look in Google Search Console and I’ll navigate to the specific page, assuming it’s an existing page. If it’s a brand new page, you. There’s other ways you can do it, and I’ll look at all the different keyword variations right that the page ranks for, and I’ll take those and I will work those.

I’ll add an extra word or two to each of those phrases, and I’ll work them into my anchor text. If it’s a brand new page, I will look for pages that are already ranking in a refs or sim rush, and I’ll see the. I’ll see the keyword phrases that that page is ranking for, and I’ll use variations of those keyword phrases to work into my anchor texts, making sure I change them every time.

And they’re not overly optimized. I, I’m not a huge fan of links. You know, some people vary their anchor texts by completely under optimizing them and saying, click here. Or use naked URLs. Or naked URLs is when I just looked to the URLs. In our, in our, we actually looked at this in our study and we found that people who use naked URLs.

Actually pages that had naked URLs as anchors actually ranked higher than pages that didn’t have any naked URLs, which is interesting. Wow. It goes to my point that I don’t think there are any bad bad anchor texts that you could use. I don’t like using click here or generic anchor text. I think Google has ways of devaluing, devaluing links that, you know, don’t get clicked or don’t have relevant information.

But so I like to, I like to have. Every anchor text contain at least one keyword. Don’t have to overdo it, but that’s how I do it. I’m, I’m rambling again. Sorry. I’m an excellent podcast guest. You are? Yeah. We’re gonna first 

Jared: find perfectly Well, the problem is, is that I’m holding up for the people on YouTube.

I’ve already written a whole pages worth of question, so we got lots to get through, but that’s okay. We, we, we have all day, as they say. Yeah. So let me ask you, continuing down the anchor text theme, cause I, I love it. I. Again, we don’t talk about internal linking enough. We talk about back links more.

When we talk about internal linking, we tend to talk more about how many you’re sending. We’ll come back to that, but I love that we’re on this anchor text focus and you’ve talked about how important it is for passing relevance. So when I look at sending an internal link from one page to another, mm-hmm.

How important is it that the page sending the internal link ranks, and how important is it that that page ranks for certain keyword. And does that play a part in what relevance it sends to the anchor text? Sorry. To the internal link? Yeah. 

Cyrus: Hugely important if you can do it, but it’s not, it’s not a deal breaker.

Yes, you want to get links from pages that rank that would be ideal, but as long as they are in the same thematic ballpark, I think it’s okay. If you. If you have a paragraph of text that’s not exactly what the page is about, but it’s, you know, it’s really, really close. It’s, I I think, I think that’s good enough, 99% of the time.

Even I, I’ll even take a link from a totally irrelevant page, but I, I think that starts to look a little spammy and I think Google can sort of sense that out. But yeah, as long as they’re in the same thematic ballpark, I, I think you’re okay. And a lot of people worry about cannibalization, like, ah, okay, here, here’s my, here’s my page about cameras.

Well, I don’t wanna link to this other page about cameras cuz I don’t want to, I don’t. Pass my relevancy signals to that other page. I, I think in practice That’s fine. It’s fine to do that. The only time you you’re gonna get in trouble is if the wrong page starts ranking, and maybe you have to rearrange some of those signals.

But I would, I would to totally link to other pages with rich anchor decks. Just don’t use the exact phrase that you’re trying to get that first page to rank for. So if, if you have, you have best Nikon cameras 2023, and you want that page to rank for that keyword, don’t ring out to another page that says, best Nikon cameras 2023 change.

Be more specific about what it’s about because hopefully it has a slightly different intent and Google can rank both those pages for very similar terms. 

Jared: How does your study, I’m gonna throw a concept at you that I hear a lot. And how does your study have any insights into it? The idea of link sculpting or page sculpting?

Right. And I’ll kind of set the. I hope I don’t butcher it. Correct me if I’m wrong any, but, but this idea that you basically control every link on a certain page. You control the anchor text and you control where you send the internal link because page rank in its concept and the link juice you get to that page is kind of divided by the number of outbound or inbound links you have, right?

Yeah. And so you want to be very careful about a page, especially if Page has a lot of. You want to kind of sculpt where it sends that link juice to and the anchor text. You know, I’m just curious how important that is nowadays, if that’s important at all. Yeah. And how much internal linking comes to play, because that’s a, that’s a, that’s a strategy a lot of people use, right?

Cyrus: Yeah. This is a great question, Anne. Thank you for asking it. Pa page Rank Sculpting is an old SEO idea that really reach its heyday around 2010, 2011. Most people don’t even remember it anymore. You have to like that. Cyrus. Yeah. The, the. And back in the day, if, you know, followed a link you could easily sculpt, pay, drink if, you know, followed one link more juice passed through the other links, so people were no following their unimportant links.

So more juice would pass through their followed links and then Google came out and they, they killed that practice. No following links. Did not give you more Jews through your, through your other links. And as a result, Google started saying things like, pay, rank, sculpting doesn’t work. And people were like, oh, pay Rank Sculpting is dead.

No pay rank, sculpting skill works. Even today, 2023. It’s just you can’t do it through the use of no follow. You can still control how page rank and other signals pass through your pages, just can’t do it with no follow anymore. So one way that you can control it as we talked about is the position of the link on the page PA links in better positions are going to pass more signals.

So your footer links probably not gonna pass a lot, but the link high and tight in the first couple paragraphs. Yes, it’s gonna pass more. The other way to control it is the number of links on the page. So if you have a thousand links on your page, Each one’s only gonna pass a tiny bit, but if you can pair it down to 10 links, you’re gonna get better signals through those.

So there are, I’m a huge fan today of page drag sculpting. And I think a lot of, i I, it’s a shame that Google kind of won that narrative with the SEO world and stopped people from doing it. Back before, back before they had this machine of the. Google search console team, the webmaster and everything like that.

Listen to old school SEOs. First, but yeah, Pedro sculpting. Bob Con. I love it. Loma, right? Yeah. Yeah. Can now, can I get into the weeds about paid sculpting for a second? 

Jared: I’d love it. Yeah. I, okay. I, I have a couple more questions on it, so you’ll probably just, you’ll probably answer them. I think it’s good.

All right. 

Cyrus: Done this a little bit. So here’s a, here’s a post we’re gonna publish here in a couple weeks. You get a, you get a preview. We talked about those, those navigation links, right? You link in the navigation. Well, Google has this thing called first link priority, which I, it’s another thing we don’t talk about very much, and that is when Google sees more than one and, okay, let’s, let’s set, let’s set the stage.

You have a link in your navigation, goes to your money page. You have a link in your post beneath. Link that also goes to your money page. Mm-hmm. Does Google count those two links equally? Mm. So first link priority, which is another concept says no. Google gives priority to the first link, meaning go past page rank through both those things, but you only get anchor text from the first one.

So we set out to test this recently and what we found is that yes, first link priority is still a thing with Google. When we have more than one link on the page going to another page, Google counts the first anchor text. They will not count the second anchor text unless it’s an image. The image exception rule.

Comes into play here so that if you have a product, product image, product description, they both link, both those anchor texts will count. But you have to be careful when you link over and over again to the same page. Cuz Google is probably only counting that first anchor text. And that’s something that that’s another way that you can pay, rank, sculpt, or pair things down.

But it, it’s important to remember that that concept we’re gonna be publishing a post on that pretty. We’ll get that in 

Jared: the show notes. Who knows? It might be time to come out right when this interview comes out, so it could be perfect. Yeah. All right. Going back to that concept of page sculpting and, and kind of diving even further, you, you talked about how if you have a thousand links on one page, it’s mm-hmm.

Be passing very little link juice per se. How, but does the relevancy get diluted? Because we’ve been talking a lot about anchor text and relevancy. Does that relevancy still pass through just as much with a lot of links on a page or does that matter at all? That, 

Cyrus: that’s a great question. I don’t know the exact answer, but I would suspect less.

There’s this concept of topic weighted page rank which is just the same way that signals pass. Page rank passes through links, topicality passes through links, and it works the same way as page rank, the way they calculate it. So if you have a page and it’s about cameras and you have a thousand links, Each one, that topic, weighted page rank is gonna get diluted if you know, with every additional link that you add.

So, yeah relevance is most likely. I’m, I’m, I can’t say definitive definitively cuz I don’t know have insight into Google’s algorithm. But yeah, I would imagine that topicality and relevancy does diminish the more links you add per page. 

Jared: So we’ve been circling about the importance of both the number of internal links you.

And the anchor text you choose to use. Again, the number of internal links is directly correlated to the traffic on the page. More the better and the very, the varied ness of your anchor text is directly correlated to better traffic. Is there a, we’ve also established though, that too many anchor texts on one page, sorry, too many internal links on one page starts to dilute things.

So is there a certain number of internal. In your study that is maybe a sweet spot where we’re maximizing the number that goes to the benefit versus the dilution that comes as a result. Yeah, I, I, 

Cyrus: I’d like to say that there’s no set number. It’s 10. It’s 10. So in our data, people will say, how many internal links do I have in our data?

We found that the, the graphs rose pretty steeply up until about 10. After that, the it, that it’s sort of died down a little bit, but we strive to get, for every page that’s important that we want to rank, we strive for an average of 10 varied internal links from different pages on our site. Tens a lot.

I mean, if you’re like me, you, you know, when you write a page, if it’s a new page or an existing page, if you’re really doing your job, you might go find three or four relevant pages. Yeah. And add some links and like, oh God, I’m, I’m done. I’m gonna go have a coffee. 10 links. 10 links is a lot. And especially, and it’s especially hard to do when you’re linking to brand new content to go back and add those links.

And there’s a. I, I have, I have a way to do that, that I should talk about. But yeah, 10 links is a lot, but it’s, it’s definitely something you should do. 

Jared: Don’t let me dive, if you have a way to d, dive in, please. 

Cyrus: All right. So adding, so there, let’s talk about the concept of adding links to existing content.

Yep. When you’re, when you’re writing a new post, you’re writing new, new paragraphs, you’re writing new links, it’s all, it’s all fresh content. So a lot, there’s a lot of debate around this and. A lot of nuance, but one popular way of adding links to existing content is you, you run a search for the exact phrase that you want.

You find it in your existing content best camera, and you l you just go back and you don’t change anything about the post. You just highlight that link and you link it. Yeah. I, I, there’s evidence that that is effective but it’s not the way I prefer to do it. Okay. Because Google can see that you just added a link.

You didn’t change anything about the content. I like to rewrite or add a sentence or two of text. To wherever that paragraph is and update the old content a little bit with the new link. I’ve found over the years that that is more effective than simply highlighting something and adding a link to existing text.

Because you get some fresh freshness signals there. You get, you get a tiny bit of new content. It doesn’t look like you just did an automated linking, but that’s how I like to do it. I, I don’t like to just highlight exact match. Take the, take the effort and update the sentence, make it look like a human was in there and actually cared.

And I think you’ll do a better job. And that probably 

Jared: would also allow you to get more internal links to hit that 10 number because chances are you maybe didn’t have that exact phrase and you wanted to 

Cyrus: find a way to work it in. Yeah. And the other, and here, here is a plug for link whisper a lot of. I, I, a lot of people don don’t make, take advantage of automated internal linking tools.

Link Whisper being one of them. We had one at Zippy for a while as part of our software suite that we, we sold off. Internal linking. It’s hard, especially if you’re scaling it out and you’re working with writers that aren’t as familiar with your content. Using automation and using tools to find opportunities.

And I won’t mention, you know, many of your competitors, but there, there, there’s a few of them out there. I think using automation in internal link building is, is necessary because finding those 10 links or more for every page especially if you’re doing it at scale, it’s, it’s difficult. It’s, you may think you know what your most relevant pages are but having machines do this, that can process and understand very quickly where your opportunities are I think it’s a no-brainer, especially for agencies and folks working with larger sites.

So I’m a huge fan of using tools and automation in internal linking to help you find those opportunities. 

Jared: Well, glad to hear the internal link expert as a fan of link Whisper. Let me ask you about that though, cuz I, you, you kind of touched on these threads. I want to outline maybe a typical scenario for a website builder, and I wanna get your take on the best approach to a internal linking strategy.

Let’s say that I’m building out a topical hub, right? Mm-hmm. And so I’m, well, let’s go down the camera theme and let’s say that I’m gonna write about wildlife cameras, and I’m thinking that’s a little different than the brand. We’ll talk about wildlife. Mm-hmm. X photographer here. So you actually picked a topic.

I could talk quite a bit. We’re gonna go down, down the road of wildlife cameras, so we’re gonna build out this topical hub all around cameras and wildlife. We’re gonna probably talk about cameras and lenses and lighting and all these different things. Mm-hmm. So as I start writing this content, I start releasing an article one by one.

I write the first article, top 10 Considerations for Wildlife Photography. I got nothing to internally link to because it’s the first article that’s just come out in this topical silo right as I get to article number. I now have nine other internal links I could place theoretically. Yep. Probably not all are relevant.

By the time I get to Article 50, I’m probably able to get 10 internal links in. Right? Yeah. But what about that article that I published the first one outta the Gate, and so there’s this problem that exists for people. Like how do you manage internal links as you’re building out, say a topical hub or a website?

When is the right time to go back and add them back in? Does it need to be part of a larger article update process? Like talk to people for how they make internal linking a part of their overall strategy and their workflow? 

Cyrus: Yeah, that’s, that’s a great question and I, I’m not sure there’s one answer for all of it.

When I do content planning and writing content briefs for clients, we usually include that as part of the process, as part of the mapping, showing how the articles will link together ideally and we’re. We’re pretty generous with those links. It’s still kind of hard to add when you haven’t added the content yet, right?

And you have to go back and you have to add that. So my, my advice is to add those, start with a plan. If you, sometimes you, you don’t plan 50 articles in advance but if you can start with a plan, do that, put those links in as soon as possible. Sometimes, I mean, it’s not ideal, but sometimes, you know, you should be probably updating your content every couple years.

At most, you can make that part of that process. If not I’m, I’m not, I, I’m a link evangelist and I don’t want to say that you should drop everything else, every other part of your SEO process to update links. Then it’s the number one priority. Sometimes life and, you know, other business decisions intervene, but make it part of your cycle.

And I. You know, I, I’ve managed content teams and I work, I’ve worked with a lot of content teams and I know it’s just never a priority unless they make it a priority. And usually it’s, usually it’s the internal linking sprint where we have to go through and do this and identify the pages. But if you can make it part of your process you know, every time.

You write a content brief or you plan content, where are those 10 links coming from? It makes it easier down the the road instead of doing that, that big sprint that everybody has to be involved in. 

Jared: Yeah, no, I think it makes a lot of sense. I mean, updating old content is something that most SEOs agree you should do, right?

Like there’s more studies that support it really helping your website than not. And yet it’s another one of those things that it’s always like, for some reason we always just wanna go publish new content. And so if you can make article updating and internal linking a part of a continuous process, that’s, that’s good advice.

Yeah. You 


Cyrus: one thing I, the total side bomb hit here one thing I hate is people have related links on their site, right? Related link widgets. You might also like and I get why, and they, they put those in like, Hey, I have internal links. The problem. No one gets to the bottom of the article and clicks those links.

You probably have like a less than a 1% click through rate on those very few people clicking and, and if few people are clicking, that’s a proxy for how much value Google is putting on those links as well. You, you haven’t done your job of internal linking if you just stuck a widget at the bottom of your page.

But one thing we experimented with at Moz quite a bit is if those links are actually valuable, put ’em at the top of the page. If, if we, if you’re, if people are searching for information on canonical tags and you have your, this article on Canon canonical tags, don’t wait till the bottom of the page to show other articles on canonical tags.

Put those high and tight near the top because they’re high intent pages that people are actually probably interested in clicking it. So if you have a, a, a related links widget or something like that, I say put it near the. Make it, make ’em relevant to, to where people are gonna click and look at serve the user.

And you can see in your analytics, again, if people are visiting more pages per visit, that’s how you know you’re doing your job because you’re giving people to things to 

Jared: click on. I know people who advocate against a related posts or related link plug in widget thing, you know? Mm-hmm. Section for various reasons.

Do you think site owners should keep. On their website, even if they keep ’em at the bottom. Do you think they’re damaging or do you think that they probably add minimal value? 

Cyrus: I don’t think they’re damaging. I think they add very minimal value. You, you can’t, and especially, especially if you’re.

You’re relying them as your primary source of internal linking you’re getting the same anchor text every time. It’s, it’s just not doing much for you. So I’m not opposed to them, but I, I don’t think people are getting much out of them. Yeah, they’re right. 

Jared: They’re let’s see. I have three topics I want to ask you about.

I’m trying to figure out which one. First, let’s talk about that naked URL topic you. Yeah, I’m very surprised to hear you say that the data correlated that there was actually a little bit of an uptick by using naked U URLs. I’m trying to even think how I would work that in to my article and stuff.

Like, why, why does that matter? And I know that Naked U URLs help to kind of even out your backlink profile, but why is it a big deal when it comes to internal. Yeah. 

Cyrus: So, and to be clear, Google, Google’s official advice is not used naked URLs use descriptive anchor texts, but the reality is sometimes, especially, you know, large organizations or whatever, your CMS spits something out, you get something that’s just a url, and I think the only reason it works, why we see good traffic, Folks using Naked U URLs is an accident.

It, it balances out over optimized anchor tax. It adds some variety. And it’s, at the end of the day, it’s still a link. It’s a different type of link. So I, I don’t think the evidence supports that Google ignores naked U URLs. But it seems like it, I think it speaks to the idea that you should.

Variety in your anchor texts and not that you should be using naked URLs. So there, there’s some. We don’t need everyone 

Jared: Listen to this podcast to go out and now start including naked url internal links. Yeah, it, although we just supports variation as the key. 

Cyrus: Yeah. Although there, there is some evidence, not strong evidence, but based on Google statements and some things we see there are, you can have such bad anchor text that Google ignores your link.

It may not pass sign a signals whatsoever if you’re, if you’re just saying click here or empty anchor text you know, which happens a lot, especially in images. Remember in image, You get the extra anchor text, even if you link to it any anywhere else on your page, it gives you another opportunity to add anchor text to that image under the alt tag.

So one mistake people make is always using the same anchor text with the image as they do with their other links. Always vary that anchor text because it’s another sh shot at. Getting, getting a small relevancy signal pointed at that page. That 

Jared: was one of my three remaining questions. Ah, images. Yeah.

So you’re let’s, let’s kind of unpack that a little bit so everyone understands. Clearly when I put an image on a page, Why is that an internal link and how do I use that to my 

Cyrus: advantage? Yeah, so image on page you can link to it the, instead of alt text, it’s the instead of a link, it’s the alt text, which counts as the anchor text in our data.

So our, over our 23 million legs, we found 5%, 5% of all internal links had empty anchor text, meaning they had no anchor or no image. Alt 95. Of the empty anchor text for images. So there’s a huge number of people in our dataset that just left their image links blank. And that gives Google nothing to work with.

So you wanna make sure that you’re filling in those alt attributes every single time With varied anchor text, it’s not a huge signal. It’s, it’s probably likely that Google weighs image links slightly less than regular links. We can’t confirm that, but, There’s some indication that that’s true. But yeah, don’t, don’t sleep on it.

You’re, you’re just leaving anchor texts on the table when you leave those images blank. So fill out anchor 

Jared: text. That’s important. Yeah. How does, I’m still wondering, I still wanna really unpack the idea that internal linking comes as a result of images because how does the image relate to a link? I know that the image has a URL itself.

When you click on the image, unless it pops out in the light box, it’s gonna be popping out with a new. But yeah. Where does that play into internal linking? 

Cyrus: Well, if you, if you, so if you go in and manually add a link to that image wrap it in an AAG pointing to another page. Yeah. So it’s not so much, it’s not so much the URL of that, the image which can cause some own its own problems in and of itself.

But when you wrap the, when you wrap the image in an AAG and link it to another page, make the link, make the image clickable. Yeah, that’s 

Jared: what I wanted to hear. Okay, cool. Okay. That’s good. And. When you do that, the alt text you’re using for that image is basically the anchor. 

Cyrus: Yes, exactly. Got it. So you, and you can, you can crawl any large e-commerce site or any site really, and you’ll very quickly find images without linked images, without their alt tags, included.

And that’s just, that’s just missed opportunities. Ah, 

Jared: we see it all the time when we do an a ref’s AHF site audit. It’s like missing alt text. And yep. That’s where you see it. Yep. You see a lot of it. Okay, so question three of three on my long list here. Don’t worry, we can go further, but these are the three important ones I had to get to you.

You mentioned it earlier, cannibalization, so we don’t get to talk about that much here on the podcast. Yeah, but I love talking about cannibalization. First off, there’s always the great debate about if keyword cannibalization is even a. And then if you settle on the camp that it is one of the recommended routes to control it is through the anchor text you use.

And I’ll outline a quick scenario just so people listening can kind of get their head wrapped around it. Let’s say that we’re talking about wildlife photography, and I have a, an article on the, the Best Wildlife camera, and then I al mm-hmm. Also have an article on the best camera for. And the best camera for birds, for whatever reason, starts ranking for the best wildlife camera by accident, right?

Mm-hmm. Like, is this a problem? Do I let it go? One recommended strategy is I can control that by making sure that all the anchor text that sends juice to the bird photography camera article is not wildlife related, but bird related. That’s an example, right? So keyboard ization, thoughts on. Do you do, do you put stock in it?

How do you control it? Where do internal links play into it? Yeah, 

Cyrus: so it’s, that’s a great question. And so people are, I, I think fears of cannibalization are greatly exaggerated and people go to great lengths to avoid it when I don’t think it’s an actual problem most of the time. It’s really only a problem in those rare examples.

And, and you mentioned one where you just have the wrong page ranking for the term, and oftentimes that. That can be solved in, in other ways. And it’s a, it’s an indication that you need to fix your main page first. But, so we did this thing at Moz where we had, so Moz had, you know, 15 years of blog content.

So we had multiple pages on every, every topic. And I remember we were trying to rank better for Canonization, which is a big term in, in seo. So we had all these old blog posts. They were still relevant. They just happened to be older and we just linked every one of them together using keyword rich canonization keywords.

Now a lot of people would fear cannibalization. In that case, you’ve got a lot of, but what we found 

Jared: going on here in this this, this good, good SEO termin, Yeah. 

Cyrus: But what we found is all the pages improved in ranking. Yeah. And the intent. And Google was able to determine the intent well enough for each slight difference in page that 99% of the time they served the right page.

But then at, we also, since we linked to all the other pages, if they didn’t land on the right page, they could just quickly go to the, the correct page and click on those links because we provided them with those internal links. Now in those rare, rare instances, Simply have the wrong page ranking in the, in those cases, yes, sometimes you might need to adjust your anchor text a little bit, point more links to the page that you want to rank.

Improve your on page copy a bit things like that. But I think those, in those instances are fairly rare. E even though there is a risk of it happening, you shouldn’t let it hold you back from linking internally with, with keyword rich anchor texts. Here’s one, here’s, here’s one thing we found and the, actually, I’m taking credit for this.

We didn’t find this. The, the folks at Search Pilot have done a number of studies on this. You add, you take, you link studies, you take a certain number of pages, you add external link, internal links to another set of pages. Now the fear is with cannibalization is that the pages you’re linking from are gonna go down in traffic or give all their juice to these other pages.

So you need to be very stingy about li linking to your other pages, cuz you don’t want to cannibalize, you don’t want to give out your link juice. But what has been found in study after study, after study when you. Internally to other pages there’s a phrase a rising tide lifts all shifts. Not only does traffic go up to the pages you’re linking to, but traffic goes up from the pages you’re linking from.

And we don’t know exactly why this is, but the idea is that when you link to relevant pages, you’re helping the page that’s doing the linking. Even if you’re using keyword rich anchor text I. It’s kind of a contradictory, it’s hard to wrap your head around, but go ahead, internally, link with that, those keyword.

You’re not, you’re probably not shooting yourself in the foot. You’re probably helping yourself and you’re helping the pages you’re linking to as well. Hmm. 

Jared: Okay. Okay. A lot to unpack there. I’m glad we touched on that. Let’s, let’s see. I I thought of another one while we were, while we were talking here.

Obviously we know. 4 0 4 pages are not good. Yep. All websites tend to have a few, so it’s not like if you have a couple 4 0 4 pages, it’s gonna really hurt you. But internally linking to 4 0 4 pages does keep Google potentially crawling that page. Yep. So let’s talk about the, that 4 0 4 and if your study had anything to do with that or if you have any thoughts on it.

And then 3 0 1 redirect or redirect hops when you internally link to a page that 3 0 1 redirects to another one page on your site. And how important is that to get cleaned up or is it really not that big of a deal? So 4 0 4 s and 3 0 1 s. Dead pages, redirect pages, these kind of topics as it relate to internal linking.

Cyrus: Yeah, so 4 0 4 pages starting with those obviously not ideal and you want to fix those as soon as possible. And you. You have a link going to a 4 0 4 page you can either drop the link or redirect it with a 3 0 1 to a page that’s as highly relevant as possible. And that’s the key. If you redirect it to a page that’s not relevant Google can sense that and the value of the link just kind of evaporates.

We see that all the time with. Sites that redirect everything to their homepage, or, you know, you bought a domain redirected all of your homepage. It’s not that relevant. It’s not gonna pass those link signals because it’s, it’s not going to a relevant page anymore. 3 0 1 s the idea that, so in theory, in theory go, if you have like 3 0 1, 3 0 1, 3 0 1, you finally get to your right page.

In theory, Google passes equal. Link juice, link equity through all of those hops in practice. Google may not follow each one of those redirects perfectly. May not, may not get to its final destination. It’s usually better to have the most direct route possible. That’s said. I’ve seen many folks Undertake these huge projects to fix all their 3 0 1 s, fix all their 4 0 4 s and nothing happens.

And that’s because possibly because they waited five years to fix them and Google’s already processed and forgot about those link links signals. But it is something you want to keep on top of and do it as quickly as possible but it’s not always gonna be your highest. Okay. 

Jared: Okay. Man, as we start to bring it all the way back, cuz we’ve talked about so much, I hope you’ve had fun.

I’ve had a great time talking about this stuff for darn near an hour now. Let’s just kind of bring it back because we, we’ve, we kind of ping ponged about a ton of topics and just reset the table, if you will, with the results of this study. So you studied 1800. 1800 1800 websites, 1800 websites, 1800 domain, sorry across 23 million pages or 23 million URLs.

You had access to Google search console data, so you’re able to directly tie internal links to their actual page level traffic. Mm-hmm. And the, I I don’t wanna over summarize, so fill in the gaps where I’m wrong, but the big takeaways we can walk away with are that the number of internal. To a page does positively impact traffic.

Yes. You recommend 10, 

Cyrus: roughly 10. 10, yes. 10 unique links with unique anchor texts, going to at least your most important pages. 

Jared: And that was my second one. Anchor text is massively important for relevance, and you have to vary that anchor text, varied anchor text one across the board. We talked about exact match.

We talked about duplicate, we talked about all the nuances around it, but in general, you’ve gotta vary that anchor text across. 

Cyrus: Yeah, and there’s nothing wrong with exact match anchor text unless you have a very small existing link profile. You want as varied of anchor text profile as possible. And if you’re a small site, if you’re, if you don’t have a lot of authority yet, I would be very wary of using exact match anchor text.

Jared: Let me close out by asking you this final question. What changes, if any, to your strategy of s. Did you make as a result of 

Cyrus: these? Hmm. That’s a, that’s an interesting question. I, my first thought was, damn, I need to be varying. My anchor texts more it, you can’t re can’t rely on it. Z you know, when I was at mos, we, we didn’t have to worry about it so much because we had gazillion links.

At Zippy, I don’t have a gazillion links, so I have to be much more careful about it. The other thing, More careful about for smaller sites is using navigation to rely on internal linking. I think for smaller sites with less authority, you have to be more conscious of putting, using, directing people in other ways than your navigation using in text in.

Links, body links and using those smartly Brian Dean Backlinko, who recently was acquired by Seresh. If you look at his blog, you find hardly anything in his navigation. It, it’s almost fr I, I think it goes a little too far. His sight is kind of frustrating to navigate. But he has, he has, you know, home about, you know, like three links in his navigation.

He relies on in body links so hard and he’s, they’re great. He in, he’s got a very strong internal linking game. And if you wanna find an example of someone who does internal linking Right. Backlinko and it paid off well for him. Yeah, it did, 

Jared: that’s for sure. So, Cyrus, excellent. Excellent having you on.

It’s so good to finally connect after all these years of reading your different studies. People can find [email protected], I know z y p p y.com. We’ll obviously include a link to this internal link study that you did. Anywhere else people can follow along with what you’re doing. 

Cyrus: I, I used to always say Twitter.

Boy who knows how that’s going these days. I’m, I’m slightly more, I’m slightly more active on, on LinkedIn, but yeah, I’m still posting on Twitter once in a while, so, yeah. Good, good. 

Jared: Appreciate it. Thank you so much for coming on board. You know, if you keep releasing these, these, these studies, we might have to get you back on for another one.

So, but thank you again for this. That was great to, to go through it in depth and be able to ask you a lot of questions that I think. When people look at it, they’ll probably have in some degree or some form. So it was really great to get this all out there. Thank you so much for joining 

Cyrus: us, Jared. Thank you.

Appreciate it.

Source link