Episode 122: E-commerce SEO: internal linking with Nathan Lomax

Play this episode:

Or get it on:

• Apple iTunes
• Anchor
• Breaker
• Google Podcasts
• Overcast
• Pocket Casts
• RadioPublic
• Spotify
• Stitcher

What's in this episode?

In this episode, you will hear Mark Williams-Cook SEO with Nathan Lomax from Quickfire Digital, an e-commerce specialist build agency about SEO for e-commerce brands talking about internal linking.

They will cover:

• Why is internal linking important?

• Can you over-optimise internal links?

• Is it possible to have too many links on a page?

• Should you link to products from your homepage?

• Can you use nofollow tags internally?

• and much more!

Transcription

MC: Welcome to episode 122 of the Search with Candour podcast recorded on Friday the 30th of July 2021. My name is Mark Williams-Cook and today we're bringing you the fourth in the SEO for e-commerce series partnered with Quickfire Digital and we're going to be talking about internal linking for e-commerce sites. So, we'll be covering questions like should we link to products from our home page, how anchor text affects rankings, should we nofollow internal links, are footer links any good, is it possible to have too many links on a page? That and many more super interesting questions. Well, if you're an SEO, at least.

Before we kick off, I need to tell you. This podcast is sponsored by the lovely people at Sitebulb. Sitebulb, if you haven't heard of them, is an SEO auditing tool. It's desktop-based. It's for PC and Macs, which I hear the team at Sitebulb absolutely love. And normally at this point, I tell you a whole bunch of things about Sitebulb or some of the specific features that I found particularly helpful because I am a Sitebulb user. I've used it for many years in the agency and we wouldn't have someone sponsor the podcast that we didn't believe in the tool or service or whatever it is that they're doing. That's the point, that I'm kind of recommending this to you from a personal level.

And this sponsorship, at least in this episode, it's going to be a little bit embarrassing because in this conversation I have with Nathan Lomax from Quickfire now talking about internal linking, Sitebulb came up approximately 162 times, to the point where Nathan had to interject and mention that his podcast, his episode, his webinar was not actually sponsored by Sitebulb. And this wasn't because I was trying to specifically plug them. It was honestly because the questions he was asking me around internal linking and tooling, normally Sitebulb, in my opinion, was the best answer for the people listening and I just think that says more than I could possibly say in this kind of segment about Sitebulb. But we're going to be talking specifically about internal links and that's one of the things I think that Sitebulb is really strong with.

So, that is, it can do this audit of your site based on the crawl and it will give you really nice data about your internal linking structure, what anchor text you're using, which pages are important based on your internal linking. I haven't seen anything else that gives it so quickly and so clearly.

Here's the great news. You get a special offer if you're listening to Search with Candour, so go to Sitebulb.com/swc and you get an extended 60-day trial of Sitebulb. So, you don't need any credit card, no payment up front. So, it's completely free to test, see if you like it. And as you'll hear in this recording now with Quickfire, there's a lot of good uses for Sitebulb.

NL: Good morning and welcome to the next episode in our series of SEO for e-commerce with the man himself, Mr. Mark Williams-Cook. Good morning, Mark. How are you?

MC: Just struggling to find the unmute there. Yeah. Thanks for having me. Number four, I think it is, isn't it?

NL: Number four away. Number four. I've had some wonderful feedback from those who are attending. Thank you for all of your kind comments coming in. For those that cannot make it live, remember, this will be recorded. But for those that are able to join us live, please do put your comments in the chat and, as ever, we will do our best to get around to them.

Now, because we're starting to get some momentum here, we've already had a few questions come in. So, I'll start with those and then we'll come to any live questions as we go through today's session. And so, Mark, without further ado, today's session is going to be around internal link building and internal linking within a site.

And so, number one for internal linking, I mean from a UX or a user experience point of view, at Quickfire, we, of course, understand how critical this is, but it's also a key part for SEO. From your perspective, can you give us a quick overview of internal linking and why it's important within the SEO landscape?

MC: Yeah. Sure. So, keep it really simple to begin with. When we talk about internal linking, it's nothing complicated. It's literally just describing how your web page is linked together through hyperlinks, which is the old-fashioned name for those things we click on that take us to other pages on the web. It's what the web is made from, which is essentially documents that are linked together. Why it's important from an SEO point of view, again, is this is really the architecture around how search engines and crawlers work. It's about them discovering pages through links and we'll get into the detail about it, but how you link different pages together and where you link them reflect strongly in their algorithms as to which pages are important and what pages are about.

So, as you said, there's definitely a user experience impact from internal linkings. It's how our users get around our site. And I think a lot of people intrinsically have a feel for how that should be. It wouldn't make sense to stick all of your main pages, links hidden away somewhere on the site. We all know that, but there are a set of principles apply to search engines that might not be as obvious, which is what we're going to talk about today. But yeah, super important.

NL: Before I dive into those, just a quick question around opening in a new tab versus the existing tab. Is there any kind of right or wrong methodology here? I've always signposted people to say if it's an external link, like a PDF or something like that. which is maybe difficult to come back, go in a new tab. If it's your own site, go in the same tab. Is there any detrimental reasons for an SEO purpose to use one over the other or both fine?

MC: From an SEO point of view, there isn't any. I think nowadays people avoid new tab links. So, this isn't my area of expertise, but I believe there is a security concern now around links automatically opening in new tabs. But yeah, from an SEO point of view, makes no difference. Personally, for a UX point of view, I'm not particularly a fan of new tabs now because I think it's rare that you get an experience where someone, say, interested in a website. They click on something and then they're taken away and they're like, "Oh no! I'll never get back to that page I was reading." I think everyone's pretty good with browser controls and stuff now.

NL: [inaudible 00:07:03] the back button.

MC: Yeah. Yeah, right. Oh! If only. If it's-

NL: [crosstalk 00:07:08] get to that level. So, let's dive in, then, in terms of prioritising pages that are most important. I guess a worry for many retailers is that actually, "Okay. I've got this site architecture here, Mark. This all makes sense. How am I going to choose which pages carry the most importance?"

MC: Yeah. So, I mean, this comes down I think to, again, we've got this overlap between user experience and SEO. So, your most important pages will be usually the pages anyway that are hard to rank for SEO that you'd want high up in the architecture. So, if you're an e-commerce site, these will tend to be things like your most popular category pages. And by design anyway, you should find these are quite high up in your link architecture. They're linked to regularly.

The things I've seen maybe some e-comm sites do a little bit wrong is when you get subcategories of subcategories and you start to get, because you haven't thought very carefully about how you're going to categorise your products, before you know it, you're in a subcategory that's three or four clicks deep and, of course, that's tricky to find for a user. And again, from a search point of view, the logic from a search engine point of view is if a page is linked to from many other pages or in prominent places, it therefore is probably an important page. If a page is only accessible through one link that is only found on one other page that itself is buried away. How important can that page really be because it's not going to get that many users to it. And I think it's a really good question to think about again in terms of users, in terms of search engines.

So, prioritising which page is the most important most likely going to be your higher traffic pages, pages that are broader in intense, so category level pages generally for e-comm sites.

NL: Perfect. Now, in question around what about page rank, because this is something e-commerce websites need to factor into their SEO plans. What actually is page rank and perhaps you can just talk a little bit about it so others understand.

MC: Sure. So, page rank is one of the bigger original formulas that Google kind of became famous for. So, the thing that separated Google from other search engines was they based their algorithm around looking at how different web pages and therefore different websites linked to each other and they came up with this algorithm called page rank, which is what Google was originally based on when Google was called BackRub originally.

And all page rank is is every page that's created on the web has a tiny, tiny amount of this page rank, is assigned some page rank. And when this page links to another page, there is a certain amount of page rank it can essentially create passover, if you like, to the pages that it links to.

So, what happens is, as this graph of pages grows, page rank kind of gets lumped on certain pages that are popular that are regularly linked to. So, the thinking is, if you had a random web surfer clicking on links on the web pages that had high page rank are statistically the most likely place that they would end up. Therefore, they're probably important pages.

Now, page rank has moved on since then. Although Google has said very recently they still use a version of page rank in their algorithm and we talk more about now what reasonable surfer models, which means if you have two links on a web page and one is this big in your CSS, it has a huge link and one is a tiny footer link, we know that there's not an equal chance that either one of those links will get clicked on. The server will likely click on the most prominent, the biggest link. So, we think about that as well now when we think about internal linking, but this is essentially what page rank and its predecessors were trying to describe, which is how popular are particular pages and from popularity, it's one metric you could use to try and discern how important a page is.

So, this relates to internal linking to literally what we were just talking about, which is I don't think it's maybe useful to have discussions anymore about page rank unless you're kind of being geeky and interested in the why it works that way. It's maybe easier for marketers, for web designers just to think about what it's describing, which is that important pages should be linked to internally, regularly from other important pages. And we can talk later about there's all kinds of software and tools you can use to help you get like a bird's-eye view of how things are linked together and how search engines understand.

NL: That's exactly what I want to ask is this whole thing talking about page rank, how do I go and find out the page rank of my site to know, "Okay. These are as seen by Google. These are your most important pages"?

MC: Yeah. So, Google a long time ago used to actually have a thing called toolbar page rank, which was ... It wasn't directly page rank, but it was a zero to 10 sort of scale that they would give to each page on the web and used to have to get a toolbar and it would tell you roughly what the page rank of this website was and it would be a good way to work out which pages Google considered popular or not.

Unfortunately, it led to all kinds of problems in terms of Google essentially then gave an objective metric for people to sell links to each other. And it just became a lot more trouble than it was worth. So, they actually canned that. So, Google don't publish any information themselves on this is what the page rank of this page is or anything like that. I mean, it would be possible to calculate it yourself because the page rank formula is published but obviously you'd need to kind of crawl the whole web, which is kind of time consuming.

From an internal linking point of view and that's what I'm going to stick to because there are other tools but we're kind of straying away from internal links there. If you want to look at how your pages link together internally, there's a really neat tool called Sitebulb that's at Sitebulb.com. They do a free trial. And what that can do is it can crawl your website and it has a link report functionality, which will list all of the internal links on your site by how many times a specific page has been linked to and it will assign them in importance on zero to one, one being most important. So, it will say, "From your internal links, these are the pages that you've kind of described as the most important." And then it goes on to give you other useful information that we haven't touched on yet, like anchor text, which we can come to later. But from which pages have I linked to the most, that's a really great tool to use because it can be surprising sometimes which pages you're telling search engines are the most important.

NL: Well, a little bit off topic, but can you talk about kind of crawler bots and bots in general? Can you talk us through what they do when they visit the website? I mean, so many people have come to us and said, "Look. We've adjusted our page structure. We've maybe reviewed our internal linking profile and all that good stuff, but now we need the site to go and be re-indexed or for someone to come do a crawl so that it can be updated." Just tell us a little bit more about what the bots do and then, when they've done their work, how then we can get it back into Google's algorithm, et cetera, so we're not having to wait months and months and months.

MC: Sure. So, when it comes to search engines, there's essentially three stages which is crawling, rendering, and indexing and they don't necessarily happen quite in that order. So, Googlebot is normally the terminology we use when we refer to Google's spider robot. So, a spider robot is essentially an automated piece of software that goes around the web and clicks on every link that it finds. And its job is basically to discover new URLs, new web pages, whether that be of images or HTML pages.

The term Googlebot, though, just to confuse everyone is kind of used interchangeably with all of Google systems by Google themselves as well. So, don't anyone be too picky about how you use it. I use it just to describe the crawler Googlebot. So, Googlebot visits web pages. It follows links basically, it will go where the links take it, and it sends back lists of URLs and raw HTML to Google. So, that means it doesn't look at stuff like CSS or JavaScript, just looks at the raw. Same as if you clicked on a web page and did view source. That's basically what it sees and that's how it discovers new pages.

What then happens back at Google HQ, if you like, is they have an infrastructure which they refer to, again, normally as Caffeine and Caffeine does the rendering part, which means that it processes the CSS. So, how the website's styled. It processes, if it can, the JavaScript. So, if you have got a website that's reliant on client-side JavaScript, Google will try and do that.

And then a decision is made and it can be made as well before that rendering is done as to whether your page will be indexed. So, just because a page was crawled, crawled means it was discovered by Googlebot, doesn't necessarily mean Google will choose to index it. If Google doesn't think the page is important enough, and by important enough, that can mean how it's internally linked to, how it's externally linked to, the content that's on the page. It may not index it. So, it may not ever be included in the search results.

Hopefully, if you've done everything right, you've got good content, unique content, you've got backlinks from other sites you've linked to internally, Google will deign it worthy of indexing and then obviously, it will appear.

Google tends to manage how often it crawls and such itself, so it does smart things like if you do a whole website redesign, you'll actually see a spike in how much Googlebot visits your website because it'll be like, "Hey! Our stuff's changed here. I need to come and visit and check out how you've redecorated," kind of thing.

Google also learns things like if your site maybe goes down because it's crawling too much, it will slow down how much it crawls. If it detects your website updates content really regularly, it might start visiting a bit more. But generally, how often your site is crawled, which will determine how quickly your content gets reindexed or refreshed, if you like, when new content gets discovered is dependent on how many links you have from other websites because it's those links that are sending Googlebot to you.

So, again, remember we talked about page rank and how the web links together and that's the probability that the server will be on any one page any particular time. Same applies to Googlebot. So, we've got trillions, I don't know, some huge number of Googlebots bouncing around the web and they're just following links. So, it means that the pages with the most incoming links at any one time are where there's most likely to be the most Googlebots sort of crossing paths and therefore, if that page changes, Google's there every two or three seconds because it's come in from some other link somewhere else, whereas if you've published a page and it's only got one link from somewhere, you may only get visited once a week as GoogleBot kind of drives by.

So, yeah. Having a site that gets regularly refreshed in Google is down to you convincing Google you're worthwhile and you're popular enough to do so.

NL: Well, I love your analogy around the redecorating and okay, the house refab. But what about if you just redecorate one room and, in our case, you're just going to have a revamp of the home page, but you're not going to do all the other pages. Can you tell these robots, "Actually just do the home page and don't worry about all the others," because actually it's a waste of their time, it's a waste of your time. You just want them to recheck the home page.

MC: I mean, there was ... Well, it is, there was. They've switched it on and off a few times. You can use Google Search Console to kind of request a page be re-looked at and indexed. That puts it in a queue that kind of speeds things up. I mean, if it's your home page, generally, you will find without any effort naturally home pages are where most links go to on a website. So, internally probably every single page links to the home page because you've got a logo you can click on. And in terms of other websites linking to you, they probably have linked to your home page.

So, if you're having to tell Google despite that to crawl your home page, you've probably got problems in that you haven't got enough links and you're really not important at all. I mean, your home page should be getting crawled on a regular basis anyway. When it comes to issues of, "This page isn't getting crawled enough," that's normally when you've got a site with half a million pages and there's a specific subset of article pages you can't get Google to look at enough. Home pages are rarely, rarely a problem.

NL: Good. Thank you. Now, in terms of particularly on the e-commerce focus, I'm keen to talk about kind of products and collections on the home page. There's a big debate around, "Okay. Do we need to show a carousel of products, best sellers typically, or most typically purchased or most popular," and versus, "Okay. Here's our top-selling collections." What are your thoughts on showing products and collections on home page versus sticking them in the nav and allowing people to navigate that way?

MC: Yeah. So, it comes ... I'm going to talk again here which you probably shouldn't do but strictly from an SEO point of view because we can get into very long user experience discussions, especially if we talk about carousels. But from an SEO point of view. So, objectively, there is definitely benefit if you link to what we refer to as deep pages from the home page. I'm defining a deep page as ones that are not directly accessible from a main menu. So, this may be specific products. So, the only way you could get to a specific product is normally maybe I've gone to the main menu, click on a subcategory and then have to click a couple more times.

So, if that was the case, as we said, Google would say, "Okay. Well, this product page is of medium or low importance because we have to click a few times to get there." If we stick that link to that product on the home page, again, as we said, we know naturally the home page is the most important page on the site not because it is the home page, but because it's linked to from every page on the site and it's where most people link to. And then we're passing that authority whether we're thinking about page rank, if we want to be nerdy, directly to that product page. So, we're telling search engines, "Hey. This product page is actually super important." So, what we're doing is not artificially, but we're raising the importance and the popularity of that one particular product page, which means it's very likely to rank better than it would do.

Now, there are limits to this. You can't just put all 2,000 products on your home page. Think of it like, you've got your jam on toast in the morning. If you spread your jam, it's going to go thinly. So, the same applies with links. If you put a thousand links on your home page, the kind of equity, you've got a certain amount of equity that you can pass to other pages. So, the more links you put on that page, kind of the thinner that's going. So, the less importance you're passing on. And so, it's good again to have a sensible amount of links. And again, that's described by when you think about this random or reasonable surfer model. Even if someone's randomly clicking links, they're less likely to end up on that product page if you've got a thousand links than if you had 10 links.

NL: The old analogy [inaudible 00:25:13] we're going through, Mark, this is a [inaudible 00:25:17].

MC: So, yeah. it can be a really useful tool if you know, for instance, that you've got, maybe Halloween's coming up and you've got some specific products that you want to rank in terms of that. So, in the week or so coming up to Halloween, I might start linking to those products from the home page in an effort to boost their ranking.

People say, "Should we put best sellers on the home page?" Yeah, you could do. I mean, the reason I would look at why they're best sellers is because if they're best sellers, basically they're already ranking really well. You kind of don't need to put them on the home page. So, I might choose something that's ranking position three, four, or five, because then, featuring on the home page actually might push it up to position one, two, or three, which is going to give a big-

NL: It's a bit like the chef's supper in a restaurant where there's certain bits you want to get rid of or chef, maybe this is an opportunity to say, "Okay. These products aren't shifting as well as some of the others. Therefore, we'll put them in a most recently purchased or whatever and hopefully encourage more people to purchase." I like that.

MC: Is that why they do that in restaurants?

NL: That is a hundred percent why the chef supper is always just nonsense from once gone by, this is how we need to get rid of them.

MC: I feel so naive now. Yeah. No, that's absolutely right. So, I think it's a good way to get people to discover other products that might have potential, yeah.

NL: Mark, I don't want to prod the beast too much, but you mentioned about carousels and your love-hate relationship or maybe your love. I don't know, but you said we could talk about it for hours. Without going into carousels for hours, are carousels right from an e-comm point of view when it comes to UX or actually do you think a lazy load or something else may be better?

MC: Don't use carousels. Everyone hates carousels. If they say they like carousels, they're lying. I've seen it proven many, many, many times how they negatively affect performance and they tend to be things that people stick on sites when it's just like, "How do we show all these products? Can't think of a good way to do it. Let's stick it in a carousel." The ones that don't automatically move are slightly less frustrating than the ones that do, but the automatically-moving carousels just need to burn, I'm afraid. I'm really sorry, Nathan, if I just ... It actually occurred to me you've just published a load of sites with loads and loads of carousels on.

NL: [crosstalk 00:27:50] recently, I have.

MC: I'm not a fan.

NL: Mark's been speaking.

MC: I'm not a fan.

NL: They've got to come down.

MC: I'm not a fan personally. I mean, always A/B test anything. My thing is if someone says to me, "I completely disagree with you, Mark. I think you're wrong. I love carousels." I'm absolutely fine. Let's do a test. Let's actually test it, but generally, from a UX point of view, and again, not my specialist area, so maybe I'm talking out of turn, but I've never had much luck or love with them.

And while we're in the midpoint of the show, I would like to introduce our sponsor Wix, who has this update for you. URL customisation on Wix is now available on product pages. You can now customise URL path prefixes or even create a flat URL structure if that floats your boat. Plus Wix automatically takes care of creating 301 redirects for all impacted URLs. Full rollout coming soon.

Also, fresh off the press, bot log reports. Get an easy understanding of how bots are crawling your site without any complicated setup right inside of Wix. There's so much more you can do with Wix. You can now add dynamic structured data, upload redirects in bulk, including error notification and warnings and fully customisable meta tags and the robots.txt file. You can get instant indexing of your home page on Google while a direct partnership with Google My Business lets you manage new and existing business listings right from the Wix dashboard. Visit Wix.com/SEO to learn more.

NL: Now coming onto anchor text on a link, is there a best practice to follow when it comes to anchor texting? Do we need to worry about over-optimisation on pages or actually is anchor text super important when it comes to internal link building?

MC: So, anchor text, for those maybe that don't know, is literally the text in the link. So, bad examples are where you see links that say, "Click here," or, "Read more." And apart from search engines, there's an accessibility thing there. So, screen readers for people who are maybe partially sighted or non-sighted, will read out the anchor text of a link. And it's helpful to describe what the following page is. So, rather than saying, "Link. Click here. Link. Click here," if it says something like, "Link, download case study. Link," whatever it is, "See product range," that's a lot more helpful for sighted users anyway.

How we use the web when you scan a page if you see what you recognise as links and they say what they're going to, it becomes a lot easier to scan the page because you don't need to read the surrounding context of the text to work that out. And again, the same applies for search engines. So, we know search engines find links important. They take that anchor text very seriously in terms of describing what they're expecting to find on the next page so seriously that one of the main things that Google tackled in their first Penguin algorithm update that targeted link spam was anchor text.

So, you mentioned there about over-optimisation of anchor text. So, the kind of old link building technique and wonderful link building, I'm talking about other sites linking to you, was you would desperately try and get people to link to you with the term that you wanted to rank for. So, if you're selling Elvis t-shirts, you're asking people to link to your category and can it please say, "Elvis t-shirts?" This was one of the main signals Google used then to detect link spam, because they realised it's quite unnatural actually for every site to link to each other with these kind of commercial terms. They normally just use either link to a home page or they just stick the URL in without any anchor text.

However, that over-optimisation does not apply internally and Google's been very clear about that and explicitly said that, saying, "You cannot over-optimise anchor text internally," so within your own site, meaning if you have your category for Elvis t-shirts, you should absolutely have at least some of the anchor text where it makes sense saying, "Elvis t-shirts."

So, if you have lots of links to pages that are buttons that say things like, "Click here," or, "Read more," or, "View all," things like that, there's a chance that you can definitely increase your ranking by using anchor text that's descriptive. Again, the tool I mentioned earlier, Sitebulb, when it lists all your internal links, it can actually list them by anchor text, so you can see the most popular anchor text you use internally and it will show you where those pages are linking.

So, it's a really good ... It's one of the on-site tasks that I'll do. I was talking to a client actually yesterday about is when we do a site audit, we will look at how we can optimise that internal link structure. And again, it's one of those great things, as I said, that helps users at the same time, but super, super, super important.

NL: Really interesting here around the image link versus text link. So, imagine you've got an e-comm site. You've got products on your home page. They might be in a carousel, god forbid. They might just be four or five on a page. That place of linking, some people choose to link the image through to the product page. Some people choose the text. Some people will put on a read more button or a buy now, whatever. Is there a preference? Should it be all of them? Should it be just be one of them? Do they carry more weight than others, et cetera?

MC: Yeah, so this was literally what we were covering yesterday with the client in terms of they had lots of articles and it had an image title and then a button that said, "Read more," I think it was. And the only thing they had linked was read more. And, in the audit, we had said, "Well, apart from anything else, this title should be linked to the article because it describes what you want to rank for."

And when we're talking through the audit with the client, they said actually, "Yeah. When we were using the site on mobile, we did find it a little bit fiddly to click on the read more button," because the first thing they were doing and it was the first thing that I did when I used the site was I was just kind of madly tapping at the image, clumsily tapping at the image.

So, there again, we're kind of creeping as we do into user experience. I mean, my view is it's always good to have as make the link as easy as possible, make it as big as possible. So, I prefer, as a user, to have the whole kind of block clickable so the image and the text. Then, again, personally, I'm not a fan of read more buttons. I think there's other ways with design that you can show that something is navigable, it's clickable, however you get a designer to tell you that. But I think there's lots of ways to do that. You don't see lots of big read more buttons all over the BBC news site and stuff.

So, yes. As long as the text is linked, you'll be fine, because search engines can work that out. They will see both links, but one gives them information. Something you can do, I've got ... I guess there's no way to prove that it helps, but I tend to do it anyway is I use a nofollow attribute on ... If I've got two links to the article maybe, say, an image and the text, I will nofollow the image link.

Now, nofollow, Google has said, "You can use nofollow tags internally on internal links," to give them a hint as to maybe which pages are not so important or links that they don't need to crawl. So, because we've got a link to that page anyway with the text, I think the kind of signal we're giving there is this is the link I kind of want you to pay attention to, either one that's descriptive, not so much the the image one.

I couldn't tell you to be honest whether that definitely has an impact but it's one of those things that is applied at a template level. It's certainly in the tests I've done never had any negative impact. So, the actual cost time of doing it is very, very, very small because it's automatically applied.

NL: So, Mark. I'm always very solutions-focused and it stresses me out when it's not like, "Well, hey. It's a solution to the problem." So, I'm a little bit curious in terms of internal linking. Is there just like a set number that you say, "Nathan, you need to be making sure there's at least this amount internal links on the page. If there's less than this, you're going to get penalised. If there's more than this, then you're just wasting your time." Is there like a holy grail number or actually it's just whatever is relevant to the site?

MC: Yeah. It's the latter, really. I mean, as I said, you'll never get kind of penalised for internal linking. I think you just need to be sensible. It's not sensible to have 5,000 links maybe or on one page. I've seen a page with 5,000 something links on. It's not particularly sensible.

NL: [inaudible 00:37:57]. I can just imagine my dad tuning in and just next time I see him, he's built his website and he's got a thousand links to the home page and I'm like, "Oh, sweet Jesus. What's coming here?" Yeah.

MC: So, it's just a matter of, again, just balancing up the user experience because essentially what we're doing with SEO always is try and deliver a good user experience but make that understandable to search engines. And sometimes, the principles when we're catering sadly to search engines make us more aware of what we should be doing for users anyway. So, I think anchor text is a really good example where people have a model in their head about how they think they're trying to be helpful even though they don't like it or use that themselves.

So, for me, carousels is one, but the read more buttons is like another one. People generally will browse the web fine without read more buttons. They don't need them and they know they don't need them, but then when it comes to them having input on how a website should be designed, they're like, "Okay. Now, it's cool. We should have a button that says, 'Read.'" "Well, do we need that, though?"

And then, you get to this conversation about anchor text and good design and that's all we're doing, really. So, yeah. The common sense does apply.

NL: Mark, for retailers listening today and any business owner really, let's say they've just gone through a site migration and they're now looking for broken links. And this seems to be the bugbear for many business owners out there. That, actually, they go through a process of migrating a site and actually they don't realise that there was a load of internal links perhaps they've forgotten about or didn't know existed and now a stack of them are broken. How can you scan the site to see, "Okay. Yeah. I've got these broken links and now I need to go and fix them"?

MC: Again, I would actually lean on Sitebulb again. So, Sitebulb, part of what they do is a full crawl of the site and they have an engine that kind of looks for problems for you and one of them is broken links. There's quite a few tools now. I mean, Semrush does site calls. You've got ContentKing. You've got Screaming Frog. So, loads of tools will do these types of scans. Specifically, what you're talking about there in terms of site migration, what is important is doing that crawl and getting a list of links before you migrate as well and actually then re-crawling links that should be redirected after the migration is super important as well as then, yes, once you've migrated, you need to, at the bare minimum, make sure all the links on your site's still working. So, if you've migrated a site and you're crawling it and the internal links are broken, you've done a really bad job with that migration.

NL: Shame on you.

MC: Failing that, you should be sad. Yeah.

NL: No, I think Sitebulb should probably sponsor today's episode because they've had a few shout outs, but just to make sure everyone's aware, Mark is in no way affiliated with [inaudible 00:41:08]. You use it, right? But at the same time, you're not getting paid to say that. Generally is a cracking tool.

MC: Yeah.

NL: That's what we need to do. Now, when we talk there about redirects, I'm interested to learn the difference between the 301, the 302, the 303. The list goes on. There's redirects for every occasion. Perhaps you can just share with the audience that the difference between a 301, 302, when to use it, when not to use it, et cetera.

MC: Yeah. So, those are what we call HTTP response codes and ones that start with three are normally to do with redirects and, yeah, the two main ones that we use are 301s and 302s. 301 is what's treated as a permanent redirect and a 302 is treated generally as a temporary redirect although there is a 307 which is technically more a temporary redirect but for all intents and purposes, it's kind of the same thing. Generally people use 301s and 302s because they use 302s more because they are the ones that have been tested more.

So, we know Google treats 302s, 307s the same, but why rock the boat when you know something's been working? So, how we use them, 301s, this is going to sound really obvious, but you do what it says on the tin. So, if you have a URL that is permanently moved, you use a 301 redirect, permanent redirect. And what will happen then-

NL: Right, Mark, like a .HTML or sometimes at the end of URLs they've not just a standard dot coded UK, right? It can be HTML or something else and therefore that would be a case for a 301?

MC: Yeah. Absolutely, yeah. So, whenever you want someone permanently to be somewhere else is when you use a 301 redirect. So, use cases might be you mentioned like migrations earlier. So, we now, URL A is now at URL B. We want it to be there permanently, so we use a 301 redirect. What that will trigger search engines to do is actually replace the URL in the search results. So, if they've indexed URL A and then you are redirecting to URL B, what will happen normally after a few days is search engines will say, "Okay. Well, they're telling me it's permanently moved. So, I'm going to remove URL A from the index and just send people directly to URL B," which is why, if you change domain, for instance, the new domain will slowly appear.

The case you've mentioned has another really interesting one. So, internal redirects, which are redirects inside your own site are quite common. So, an example might be whether or not your URLs have a trailing slash on the end, because annoyingly, the way search engines work is they treat a URL with and without a training slash as two different pages.

So, generally you will pick one kind of that you want to use. So, you're going to say, "All our URLs do or do not have a training slash." And then, you will set up a 301 redirect that pushes, say, all non-slashed pages to slashed pages. So, you're squishing those pages together so you're preventing duplicate content and you're doing that permanently. So, that's another use of a 301 redirect.

302 redirects, guess what, therefore when URLs are temporarily moving but the use cases are quite limited. Essentially what the advantage is is that if you use a 302 redirect, so if URL A temporary redirects to URL B, search engines generally will not update the URL they're showing in the search results. So, if you are temporarily showing, for instance, you're sending people off to a page because of coronavirus maybe but you want to keep your old page ranking, you might use a 302 redirect if you know it's only going to be like a month, because you can send users there very reliably and search engines will keep your original page ranking. So, when you remove that 302, it would just go back to what it was.

Word of warning, that if you leave 302s in place for months and months, they will just get treated like 301 redirects. Search engines will say, "Hey. They probably used the wrong one. It's been six months. It doesn't look like we're on a break. It's actually happening, so I'm just going to treat it as a permanent redirect."

NL: Love it. Two questions on that. One is around 301 redirects. So, let's say you go through the site. You make a load of 301 redirects. You're really chuffed and then, oh, crumbs! You've made a mistake. Actually, you look three days later and those 301s you've set up, maybe a couple of them aren't going to the right place, et cetera. Can you just re-upload and go again or actually do you need to give it some time for Google to correct itself? You don't want to confuse it by saying, "Go here. No, no. Go over there. No. Made a mistake. Actually go over there."

MC: Well, first of all, you've done a bad job and you should be sad again. Secondly, if you have ... So, say you have migrated URL A to URL B and then you've gone, "Oh, crumbs! We should have migrated it to URL C." You absolutely should just update the redirect as soon as possible. So, A goes to C. And then your B URL, if that's also meant to go to C, that should be updated as well to go to URL C. What you don't want to do is create your ... What I'm trying to avoid there is creating redirect chains. So, say A went to B and then you realise that it's actually meant to go to C, you don't want to then just redirect B to C, because then, if someone went to URL A, they'd get redirected to URL B and then the server will go, "Oh, no! Wait. Carry on. You've got to go to URL C." And this is going to cause latency for the user and Google stops crawling after five redirects. So, there's a hard limit for Googlebot.

And that sounds like a lot, but actually, and I've seen this happen when websites over the years just have these redirect rules get built up. You've maybe done two site redesigns. That third one, it's not uncommon then if nobody's looking after it for these redirect chains to start appearing. So, maybe it's going from HTTP to HTTPS and then they set up a inefficient redirect that then does the slash to the non-stash and then A, then B, then C. You're already at five or six redirects.

There's a really good example with, I think it was Topshop online or, no. It's House of Fraser. Sorry. So, with House of Fraser, they lost millions in online revenue because they ended up creating a big redirect chain when they were just moving from non-secure to secure and they lost half their search visibility within a week. And this was millions of visitors. So, yeah. Bad things can happen with simple things if you don't pay attention to them.

NL: Question with ... Is it relative linking essentially when we're not putting HTTPS in the whole link when we're doing internal link? We're just going from the forward slash. Perhaps we could just share a little bit about that for those that are looking to create internal links within their pages.

MC: Yeah. So, you've basically got absolute or relative links. Absolute links are ones that specify the whole address. So, as you say, they start with the protocol, normally https://www.mysite.com, whatever is, forward slash category product. Relative links are on a page and they are relative to your current kind of position, your visit. So, browsers and crawlers can work out that you're on this site and the link just goes to forward slash category product.

Generally, most CMSs will do this automatically. Absolute linking is much preferred just because there's no room for error there. When you start using relational relative linking, you've got room for errors to creep in, whether it's with crawlers or your own internal linking, you're just making an opportunity for things to go wrong and generally, good design practices, you're trying to create systems where there is no opportunity or there's the fewest possible opportunities for things to potentially go wrong. So, there's no benefit to relational linking. So, there's only a potential risk. So, it's something I never do.

NL: Folks, just to reset the room, we've got about five minutes left and we're going to keep going through as many questions as we possibly can. So, please do keep them coming in. This is your chance to ask Mark anything around internal linking and SEO in general.

Now, my next question comes to orphaned pages on a website. Let's say we're using some pages for a Google ad campaign perhaps. What do we do there around internal linking? Do we just avoid these because essentially their purpose is just as a landing page to direct relevant traffic to off the back of an ad?

MC: Yeah. So, orphan pages, again, just give a definition to everyone, are pages that are not linked to from any other page on your site. So, generally they won't get found. They won't get indexed. With Google Ads, I try and encourage clients when we do Google Ads to use the page that they would want to rank organically, generally because I'm like if you're paying to get people on this search traffic, we would like a chance to rank organically as well. Sometimes those pages, of course, are slightly different or you might want to A/B test things. If it's the case where you have an orphaned page, if you like, and because you're testing a new PPC landing page, what I tend to do is you can make the orphan page. And then I use a canonical tag back to the kind of other variation that is crawlable, and this means that if you get any links to your Google Ads page that you get the benefit for them as well.

So, the tricky thing is, if you have an orphan page and you get links to it, it will become indexed. So, people could potentially just land even though it's not linked to on your site, it could land straight from Google on this page, unless you specify to Google as well, "It's orphaned and we don't want it indexed." And then you've got the potential downside of if it's not indexed and it's orphaned and people do like it and link to it, you're not getting the kind of halo benefit from that which could help the rest of your site to rank.

So, I would say that maybe isn't like a correct fits-all answer and it will depend on what you're trying to achieve, but there's lots of options available when it comes to, if you're specifically doing PPC and testing different pages outside your internally linked pages, but you just need to be aware it is something you need to actively have a plan for and manage again or you're just going to be wasting potential benefit you could get.

NL: What are we doing about footer links? A lot of websites have kind of useful links in the footer and typically links to Ts and Cs and stuff like that. Any do's or don'ts when it comes to footer linking?

MC: Yeah. So, footer links are really interesting because they used to work really well because again, following our line of conversation earlier, their link ... If you put links in your footer, they're linked to from every page. So, they tend to be important and generally you've got a little bit more space to use the anchor text that you'd like.

However, Google's pretty smart now in that it will recognise what are menu links on a site. So, Google will work out what are templated links and what are kind of editorial in content links. And certainly Google knows the position of those links on the page. So, again, going back to our reasonable server model, links in your footer are not going to be anywhere near as valuable as links in your main menu, assuming you have a main menu that's easily accessible at the top of your page and assuming your footer links are kind of tiny links tucked away in the footer.

I think they can still be useful, primarily for that kind of anchor text stuff. Sometimes, I use them as a shortcut to fix other user experience issues. And I would put my hands up here and give our own With Candour site as an example here, which is that we designed the site, put it live. And I didn't use footer links, but one of the things we did on the home page was basically I put a list of some of our key services on because, when we were actually doing user testing and getting and looking at analytics, it was becoming apparent that, while our menu in the main menu was logical, it didn't necessarily just kind of tell people what we did and what services we offer.

So, in the end, I just opted to, "Well, let's just use some home page space to list the half a dozen main things that we do for people." And it becomes very clear then to visitors when they scroll, "Okay. Well, this agency does these things, but then I now have nice deep links to those pages from the home page." And that's one of the reasons why a lot of those pages rank quite well because we've got those links.

So, when I'm using footer links, it's normally to kind of fix other sometimes sort of user issues, so they can be helpful, but they're certainly not like a hack that's going to make pages rank well that otherwise shouldn't.

NL: Mark, last couple of questions. We've just had some lovely feedback in from Mahala, which is really the theme of the last four sessions. She says, "I can't believe this level of practical knowledge packed into a live chat. Thank you so much." It's really lovely to hear that and I'm glad you're enjoying it.

Two to go, Mark. We'll eke a little bit more value out of you. In terms of any other tools you recommend when it comes to internal or external link building, are there any tools that we can give to the audience today to, say, from 10:00 when we finish this session, go away, go and look at Sitebulb, go and look at this one, get yourselves an account and get cracking?

MC: Yeah, absolutely. So, off the top of my head, apart from Sitebulb, Little Warden is a really nice tool for monitoring the mundane, I think, because they they put it. So, this is if things break, it's a really, really good tool to monitor if pages are going down, links are breaking. That kind of thing.

Screaming Frog has been around for a long time. One of the most popular web crawlers. It's similar to Sitebulb but different, but another tool a lot of people use.

ContentKing, so you can go to trycontentking.com, get a free trial. That's a cloud-based crawler for your site, so it's basically like a software as a service version I'd say of something like Sitebulb. So, it will constantly crawl your site. You have to pay monthly for it, but the advantage is, you can get very fast updates if something breaks like Little Warden.

And then, you've got the kind of suites of tools like probably Semrush is one of the most widely known ones, very widely known because they had their ... I think they're the only SEO tool. They had their IPO a few weeks ago and I thought it was quite interesting. So, it's the only SEO tool that's has been through that process. So, everything's becoming very serious now that we've got tools publicly traded. But that does a whole bunch of stuff, but that's got kind of site audit, if you like, facets similar to ContentKing and will give you alerts and will tell you about anchor text, that kind of stuff.

Yeah. There's loads of tools out there. It really just comes down to your budget and how big your site is and if you've got people to actually use them. As with any tool, without pressing my circus of analogies, these tools won't do the job for you. They are that. They're tools. They will tell you things. Some of them will give you false positives. Some will completely miss things. They're all limited to what they can do in comparison to a human. And the only value you get from them is when you intelligently apply what they're telling you, because I've seen again. Lots of people buy tools. They look at them and they don't really do anything, don't action it. But there's some good ones to start with.

NL: Mark, amazing. Thank you. So, just recapping, we've talked about jam on toast, we've talked about Elvis t-shirts, we've talked about HTTP response codes, we've talked about Sitebulb for the 10th time. We've talked about all sorts of tools and it's been a cracking session, but for those that are perhaps catching the very end of this episode, what is the one thing you would encourage everyone to do from today when it comes to internal link building?

MC: Yeah. I would just, whenever you are ... Well, I'd encourage people to whenever it's relevant on any page, link to anything else that you mentioned. So, if you mentioned product services, categories, link to it and think about that link in isolation. Is it describing the page you're sending people to? It's just a really nice usability, accessibility, SEO thing you can do. So, just don't have click here read more. If I can reduce the amount of click here and read more on the internet, then I'll be a happy person.

NL: And carousels. Back off the carousels.

MC: I'll leave carousels alone, for this is my thing. This is, yeah. That's another battle for another day.

NL: Now, folks, Mark has very kindly once again given up an hour of his time to share all this knowledge with us today. Mark, for anyone that's interested to find out more about yourself or the business or to start working with yourselves, how can they find out more and where can they go?

MC: So, I'm pretty active on LinkedIn, so you can find me, Mark Williams-Cook. If you just Google Candour, you should find the Candour Agency. If you do, if you're interested in talking to us about SEO. And also, I run a weekly podcast called Search with Candour, which is 20, 30 minutes every week just talk about SEO, PPC updates and such. I see there's just one question unanswered on the end there, which is-

NL: [crosstalk 01:00:33]. Yeah.

MC: ... "Is there any SEO certification you guys recommend?" To be honest, the answer's no to that. There isn't any particular certification. Certainly, as an employer doing SEO for 16 years, that I look for or think, "Oh, that's really kind of worth it." There are courses that are probably worth doing, but there isn't a certification that the industry looks for. No.

NL: Perfect. Mark, as ever, it's wonderful to spend the last hour with yourself. Thank you for providing so much knowledge and wisdom. Thank you to all those that have joined us today and we look forward to seeing you next time. Take care.

MC: Hope you enjoyed that episode. There's more. There's always more. If you would like to join in on our LinkedIn live sessions, we will be doing another SEO for e-commerce series and we may actually extend this series because it's been so popular. Opening up every time we're getting more and more questions, but if you follow me on LinkedIn. So, just search for Mark Williams-Cook, you will find me. I believe I'm the only Mark Williams-Cook there. If you find me and either connect with me or follow me, you'll get notifications as to when we're planning on doing these live events and it'd be great to have you join in.

Of course, I'll be back in one week's time with a normal show for you, a kind of regular news, SEO, PPC episode which will be Monday the 9th of August. Hope you tune in. Tell a friend about the podcast if you're enjoying it and, of course, I hope you have a wonderful week.