Episode 117: E-commerce SEO live Q&A part 2 with Nathan Lomax

Play this episode:

Or get it on:

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

What's in this episode?

In this episode, you will hear Mark Williams-Cook answering live e-commerce SEO Q&A with Nathan Lomax from Quickfire Digital, an e-commerce specialist build agency about SEO for e-commerce brands.

They will cover:

• Best practices for products in multiple categories

• How to handle out of stock items

• Are there optimal lengths for descriptions of products

• How should you use variants of keywords

• Is it okay to hide product details behind tabs?

• How do you handle discontinued products?

• And loads more!

Transcription

MC: Welcome to episode 117 of the Search with Candour podcast recorded on Saturday the 19th of June, 2021. My name is Mark Williams-Cook, and glad to be joining you here again today, to bring you part two in our e-commerce SEO series, we've been partnering, if you've been listening to previous episodes with Nathan, from Quickfire Digital, who are an e-commerce build specialists, and we've been doing an online series every two weeks talking about e-commerce SEO specifically. In this episode, we took an hour to answer your questions live on LinkedIn and share them with everyone. If you'd like to get involved, you can follow me on LinkedIn, just search for Mark Williams-Cook. I'm pretty sure I'm the only Mark Williams-Cook on there and you'll see the dates that we're doing these events.

We're running them every two weeks at the moment. So if you are working in e-commerce or even if you're working in an agency and working on e-commerce sites, and you've got questions about SEO, you want to drop by and ask, I'll do my best to answer them with Nathan for you. I hope you enjoyed this recording. Any comments or questions, as always just drop me a line on social and I hope you enjoy it.

Before we go to today's recording, I just want to tell you, this podcast is very kindly sponsored by Sitebulb. Sitebulb is a desktop based Windows and Mac SEO auditing tool. Incredibly helpful if you do have e-commerce sites. One of the many, many features that comes in very helpful with Sitebulb, is they have a really cool tool that can check your structured data.

And this is something we're going to talk about in the recording, which is that, so many e-commerce sites, maybe half the e-commerce sites that I come in contact with, especially SMEs are still missing big opportunities with schema, so basic, basic things like, is their product and images marked up with schema. Is it labeled for search engines? The great thing about Sitebulb is, you simply check a box and it will run these reports for you. And it will also check for the schema you do have on your site if it is working. Because again, in so many cases, you'll find it is just actually broken.

Sitebulb is completely free to try. If you go to sitebulb.com/swc, you get an extended 60 day trial of the software. You don't need to put your credit card in or anything like that. So there's no obligation. Give it a go. I'm sure you will absolutely love it. So from us and Sitebulb, hope you enjoy the recording.

NL: Good morning and welcome to part two of our SEO for e-commerce series, with myself Nathan Lomax, co-founder of Quickfire Digital and our special guest Mark Williams-Cook from Candour. Morning, Mark. How are you doing?

MC: Morning Nathan. Good, thank you. Thanks for having me back.

NL: No problem at all. A great first episode. I know we've got quite a lot of people tuning in today, over 150 people registered to join us this morning. For those that are new to this format, we try to put these on as frequently as possible. It's a real chance for you to ask those burning questions you've got, in this case around SEO for e-commerce. If you are with us this morning and are joining us live, then please do jump in at any time, ask your questions and I'll do my best to answer them. Now, for those that perhaps have been here before, you'll see that we get through as much as we possibly can. I've actually got some questions already. I'll start with those, and then as we go through today's session, if you want to ask anything or you're unsure, jump in, get involved and ask Mark and I anything you would like.

So Mark, without further ado, let's get started. Last time we talked a little bit about SEO across the site, but today we're really going to focus on product pages. And so e-commerce product pages are arguably the lifeblood of a business, let alone a website. Where do you start with SEO in mind when it comes to product pages?

MC: Wow. There's loads of things we could talk about for product pages. A few things I think I've seen that catch people out is, firstly, categorisation of products. To give you an example, let's say we have an e-commerce shop and we sell furniture, right? We have our product page for a basic wooden chair that we're trying to sell. Okay. That might fit into a couple of categories, so we might have to say dining room furniture. So that fits in that category. But also it works maybe as, I don't know, outdoor furniture. Okay? This makes perfect sense from a user point of view, makes perfect sense from an e-comm manager point of view. So you go ahead and you put your product, you tick yes, in both of these categories. Everything's great.

Now what you've ended up there in a lot of cases is, a situation whereby you'll have a site and you'll have your site name, Marksfurnitureshop.com, whatever, forward slash, and then you have the category names. The way we've just set it up, we might have furnitureshop.com/outdoor furniture/ and then our product page wooden chair. And then we'll have another URL that will be our shop.com/dining room furniture, slash our product name. What we've done is we've created essentially two URLs for the same identical product. We've caused an issue there, in that search engines are going to get confused, because if someone does a search for that product page, it's like, there's two here or three or even more, which one do I rank? Firstly, having a plan around that; if you have products that do fit into multiple categories, one, the most simple solution is something like a platform that Magento does out of the box, which is to have URLs that basically don't include the category in the slug.

So you just have a shop.com forward slash product forward slash wooden chair. So then you can put it in all the categories you like, and you don't have to worry about creating duplicate pages. That itself creates many solutions, other issues around breadcrumbs and stuff. But we'll talk about it another time. Other things are, well, I always bang on about schema, so I'm not going to pass up an opportunity to do that. I still see loads of e-comm sites going live without schema. So schema is that thing to label your data for search engines, there is a specific schema that is great for e-commerce sites or product schema. And that will give you the opportunity then when your product pages are ranked in Google, it will get these extra bits of data that says it's in stock and it's 39.99.

So really, really important. There's other benefits to having schema, but that's one of the main ones and probably 50% of the e-comm sites that come to us to talk about sci, just have a quick look at their product pages, they don't have schema. Lastly, I'm just picking three here, because there's many, plans around discontinued products. Again, most e-commerce businesses we speak to, the ones that have some turnover of product, meaning a product has a season or it has a shelf life and that it's discontinued, they just delete the product when it's discontinued. Because again, that seems the obvious thing to do for users and for the site. And this is where SEO comes in, right? We need to make this a friendly thing for search engines too.

If you have a product page and it is ranking in Google for something, if you just delete that product page, obviously it's going to be a 404 page, not found, search engines are not going to want to rank a broken URL, so it will get dropped and you will lose whatever traffic you are regularly getting from that. One of, and this will depend on your situation, but one of the many potential ways around this or ways to get maximum opportunity from this, is to make a discontinue page to say, okay, so you leave the page title the same and you say this product is discontinued. And then you can actually maybe point them to the nearest possible product. Maybe this is the new model of this, for instance. Because even if say our model is replaced, if it was a popular product, you would likely still have this hangover of searches happening for that product.

And you don't want them to go to a competitor and you're still serving the same intent. Maybe I'm looking at this printer model for instance, that you sell and then you, rather than just not appearing and someone else appearing, you can say, okay, no it's been discontinued, but here's the new model. You retain that traffic. Again, a lot of businesses don't have a plan for what to do with discontinued or underlying products.

NL: Let's just revisit that a little bit more Mark. Let's say it's a Christmas promotion and you've got a Christmas category and then Christmas products, for example, come January, they're no longer relevant. Then you just change that page, remove the product. But instead of officially removing them, you just say, look, these are now out of stock until next Christmas, please check out our upcoming Easter range or something like that.

MC: Yeah, I think seasonal ones are always interesting. I've had lots of discussions about product pages, about categories, about even bits of content that are seasonal and wherever possible is beneficial to keep your hours alive in search engines. Again, you've got that divide of, you could for instance, like you say, remove all the internal links to those seasonal categories. You can no index the pages just to tell search engines in the meantime to actually not show them a search results, but that would be certainly more beneficial long term than deleting a category or making a new one the next year. And again, it just comes down to, again, what's going to work for the user, because it certainly doesn't make sense in the middle of June for them to be landing on Christmas offers of course.

NL: Yeah. Makes perfect sense. Justin Reid has just asked a question. Yes, this will be recorded and we'll make sure that you've got access to it through the group that we've created. If you're not part of the SEO for e-commerce group or events that we've created, get yourself involved. I'll happily send you the link and then you'll see today's recording and also get access to the previous recording and future recordings.

So Mark jumping on to the next question here, which is around product descriptions and this is a topic that's always brought up when we're discussing web builds with people, is it right to stuff them full of keywords and variants or actually is this something to avoid and is there an optimum length or an optimum presentation for these descriptions?

MC: I feel like it's a loaded question. Do we stand for the keywords? Let's start with Yeman. The short answer is there's no optimal length, right? For a product description. You see bits of advice floating around and saying, it needs to be 200 words or something. Right? Of course there is a correlation between product description length, and generally how well a page ranks. If you try to write all product descriptions in three words versus writing them in 300 words, yes, you would expect to have the 300 words rank better because it's very difficult to describe products well in three words. But that's not because of the length of the content. That's not why it's ranking. It's ranking because it's more helpful and it has the information that people are searching for.

The optimum length is basically the length that it needs to be, if that makes sense, to get the relevant and best possible information across. In terms of actual, just length of pages in general, the amount of information, I've always been a fan of basically as much as possible. So Amazon really were the people that pioneered this. If you zoom out on Amazon product pages to see the total amount of information they've gotten there, they look like strips of sellotape, because they're so long and they have so much information contained in there. I've heard arguments previously saying, well, we need to keep it short and sharp and keep people on the, we want them to add to the basket. But in every test I've seen, in every user test, in every constitutive test looking at analytics and AB testing, is that the longer, more detailed pages outperform the shorter ones.

It's not that people are getting distracted like, I got so caught up reading this detail and I forgot what I was trying to do, it’s that the longer content answers the questions that overcomes objections. And from an SEO point of view, of course, if you start including, take our printer example, details about all the different types of connections it has, or if it's a portable printer, what the battery life is, you can then capture searches for that specific information. If someone's searching for a portable printer with at least two hours of battery life, and you've mentioned that, then there's a chance of you ranking for that. If the information is not there, then of course you won't. And lastly, on your first part of the question, which was, do we put loads of keywords and variants in?

I think the question is why would we use variants? By variants, I mean another way of writing a search, and the answer is, because not everyone does the same search. Words don't mean the same things to everyone. Even native speakers, especially even the UK, in different parts of the country use different words for the same thing. As part of the more in-depth content analysis, you would look at, are there meaningful variations about what people are searching for? And the second bit of that puzzle is, do search engines already understand what those variations are? To give an example, if you have a variation on a type of product, for instance, if you search for them, both in Google, sometimes you will get pretty much identical sets of results. Okay?

And that tells you that the search engine understands that the person's looking for the same thing. So then my priority, should I include that keyword variant in my copy or whatever is low. It doesn't need to be in the page title, for instance, I might mention it somewhere if it makes sense to do so. If you have a variant that's so different, or just one of the edge cases where you type both into Google and you get two different sets of results, then that's telling you actually, for whatever reason, the search engine doesn't really understand they are the same thing. For instance, an interesting one I've been working on recently was looking at coach hire or mini bus hire, which are kind of the same thing. And looking at the overlap of sites that Google ranks between those two such terms.

If they are different enough, then yes, you need to include those variations. And then you fall back on your keyword data for which has the most searches, maybe if you've got the data, which converts best, if you've got PPC data for instance, as to which one you lead with. Again, it's very much balancing the user need versus you're plugging the gaps in the technology that exists, which is a lot of what SEO is, right? It's just plugging technology gaps that search engines aren't quite as smart as people yet.

NL: Yet.

MC: Yet. Yeah.

NL: Mark, I've got a first question come in from our audience today which references our last session. And for those that have perhaps just joined, just a reminder that we have done one of these before, a slightly different topic and that's available within our LinkedIn channels, both through Mark's channels and my own, but if you can't find it for whatever reason, please reach out. But last session we did touch upon the big opportunity within image search, particularly for certain types of e-commerce businesses. This is a massive opportunity. What should you be looking at with your product images to stand the best chance of ranking?

MC: Again, there's a whole separate branch of SEO really, deals with images and image search. One of the things we touched on to frame this for everyone is, a lot of e-comm searches actually start within Google images, especially things like fashion. I do it all the time, right? If I want a new pair of high top trainers and I decide I want blue ones or something, I might type something like, blue high-top trainers into Google, but rather than try and see which websites are ranking, you go into images, it's Google image search, and then you just start scrolling through until you see something like, and then you click on it. And then you go to the site where it is. That's repeated a lot in e-comm, because shopping is a visual thing. So there is a big benefit to making sure that your Google image game is up there.

The basics are, and again, as we covered, if you can, avoid using stock imagery. Again, if you look at Google images, you won't see the same image repeated over and over again, and that's because it wouldn't be helpful to do so. If you've got stock imagery and a hundred other sites are using it, and you're all targeting the same key phrases, it's unlikely that you're going to rank. Other basics are, it's more for accessibility, but you've got the opportunity to give out text for an image to describe the contents of the actual image. Google does try to extract information about the subject matter of the image, from the content of the page, those includes captions, image titles. So wherever possible you make sure images are placed in with relevant text on pages that are relevant to the image subject matter.

URL structure is actually one of the things that impacts image ranking. Google uses the URL path, as well as the file name, to help understand the image. And that's a little bit different to web pages, but it's worth considering organising image content, so the URLs are constructed logically. Schema, again, of course, can't answer the question without mentioning schema. Google images supports product schema, meaning, again, you can in Google images, it can start saying in stock and show that that thing is for sale. Again, basics, make sure the images are optimised. So that means that they're compressed. So they're small, they're fast to load, they're responsive images, they work on different devices.

If you want to appear in Google discover, so for those that haven't heard of it, it's part of the Google app pit, it guesses the thing you're going to be interested in next, surfaces articles for you. They want very large hi-res images as well for Google discover. What else? Image sitemaps, so you can actually, apart from a site map of your pages, you can actually generate a sitemap of images to make sure Google understands where or finds all your images and can categorise them a bit easier. So you can do page site maps, video sitemaps, image sitemaps as well. I think that's everything off the top of my head. It's quite a list.

NL: Yeah, yeah. I hope everyone's got a notepad to hand. Nadeem, I'll come to your question very shortly about what makes content useful. But just building on from what we've just talked about Mark, what about videos? How about product videos, similar processes, or actually a whole nother world of oxides? Is there a video schema or is that supported by product schema?

MC: Yeah. A lot of the things I've just mentioned apply, in that you want it in relevant places, blah, blah, blah. Yes, absolutely. There is a video schema as well that can help describe what's inside videos for search engines. The main difference I would say is, you've got to think about YouTube as well. YouTube is one of the largest search engines in the UK. And if someone knows they want video content, a lot of the time they will, again, just go straight to YouTube. It's like a whole different channel. And then you've got this rabbit hole of YouTube SEO. So how do you make your video rank well on YouTube? Which is very different from Google. You mentioned schema, again, yes, you see the searches that will really interest you are the ones that trigger the video results in Google.

It's not normally actual product pages unless they contain things like how to, so how to use products or how to guides. You'll see a lot of the time Google will surface videos at the top of the search result. And again, that's indicative of the intent of the user in that Google says, okay, well, I understand that this search is best served with a video result. Obviously if you do not have a video, guess what, you won't rank for it. Video as a whole, I would say is nowadays super, super helpful for e-commerce, anyway, from again, from a user point of view. The technology is still chasing video, in my opinion, in terms of search. We're seeing big leaps, you can upload a video on YouTube and it gets automatic captions. You will now see if you do a search in Google and it returns YouTube video, it has the recommended places to start in that video to answer your specific question.

And that's amazing when you think about it. Google is like, watch the video for you, and you've asked a specific question about how do I change the battery in my portable printer? It's like, yeah, three minutes, 20 seconds in this video is where you want to watch. It doesn't always get it right and that's a huge task, because videos are very big files. They're very heavy to process. My mind is still mainly on the user when we think about video. But again, from an e-commerce strategy point of view, the cost of making good video now is rock bottom, compared to what it was 10 years ago. It costs thousands to make good videos. Now, basically everyone can do it with a plug-in mic and a smartphone.

NL: Mark, just on the video piece, you mentioned stock images earlier. Is there a way that, actually, if you've got a professionally shot video versus stock video clips from Shutterstock or something like that, again, is there preference in terms of, okay, it's animation versus video or speaking to camera versus something else, or actually as long as it's video, they're all categorised in the same box?

MC: That's a really interesting question because you see in Google image search, you can filter it to things like drawings or clip art or real life photos. I haven't seen any of that filtering yet for video. I've certainly, I've never actually even heard anyone discuss whether different types of video are optimal. I think you can jump a few steps ahead of that and just ask the question, what's the most useful and that's going to be the correct answer. If you remember, I always keep in my mind, these algorithms are following us. We're not following the algorithms, right? The algorithms try to understand what is the most helpful thing to a person. Sometimes you can end up with weird answers if you try and turn the mirror back and be like, what is the algorithm telling us is helpful. Certainly you can get feedback loops, which can help guide you in the details.

But I would, again, just think about the user. Okay. We want to do case studies with our clients then, talking heads kind of make sense, right. But it certainly wouldn't be worse if it was maybe an animation.

NL: Mark, let's go to the audience. We've got a couple of questions coming in. I'm going to start with Nadeem. And just for those that are listening, that perhaps have a question, please do just pop it in the comments and I'll do whatever I can to get round to it as quickly as possible.

So Nadeem asked, what makes content useful? Which is really the million dollar question. Right? But he actually goes on to say that in his opinion, if your content is long, it means it covers lots of good things. And that the content is short irrespective of the content, perhaps it has no meaning or information. I guess the big question off the back of that is, well we might not be able to, but say what makes content useful and what doesn't, because it might be subjective.

What we can say is, on those pages, Mark, if you want to include loads of content, what are your thoughts to the read more recordings and things like that, where you can actually show quite a condensed product page, but actually for those that do want more information, you can give them that experience as well?

MC: Well, firstly, I would respectfully disagree that the length of content is related to how helpful it can be. Our SEO course has got some great examples of this. So the example I used for years was a search about when did the clocks go back in summer? And the number two ranking site had hundreds of thousands of links, you clicked and it was a regional newspaper and it had about 5,000 words. It had a video, it had a poll, it had images of people sleeping and the history of clock changes, why do we do it? And you really have to dig through to find when the clock changed. The number one ranking site was about three lines and it just said in huge letters, the clocks go back on this date, right? And this poses the question, if your search intent is to find out what date the clocks go back, which of these sites would you prefer?

The obvious answer is, I want the site with the one line that just tells me the information that I need. It's very easy to churn out long content that basically means nothing. While I think you need to be careful about the things Google tells us, because they sometimes phrase things very tactfully and there's between the lines things, they have been incredibly clear and said that the length of content is not... we don't take that into account. It's not a thing. Of course there is that correlation normally, because I've seen loads of studies that say, look, hey, the stuff that ranks number one is longer in general. And of course, because on average, people have spent more time, more effort and more love and thought into that type of content.

I think that the actual answer to the question about what makes content useful, is very simple. It's content that answers the intent of the search as quickly as possible. And that I would say, that's it. Anything else is just fluff. You see people complain about recipe sites a lot. There's that running joke that you have to read someone's life story before they just tell you how many eggs they need. Right? And it's annoying. I'm definitely a fan, and it's why we're seeing things like featured snippets in Google and why Google gives us instant answers. They're not doing it because people dislike it. They do it because the data shows people like it. To get onto your addition to that question, which is about the view more, hide more, click tabs. Again, that's a really interesting question because Google's advice on that has actually changed.

The advice used to be that, if you hide things behind tabs we will take that as a hint, that it can't be that important, because you've hidden it from immediate view. However, now we are definitely waist deep in a mobile first world. Screens are smaller. Real estate is important. I think Amazon's a really good example for this. So rather than hide a whole section away in a tab and Amazon will do something like list the technical specification and they will list maybe what they know are the five or six most important bits of information about this product. And then they just have a view with more links that then we'll expand if you need it. From a technical point of view, all of that content is always visible to search engines at all times. But it does help, it balances again, that user experience of, yes, we want all of the content on the page, but it's not particularly helpful to sometimes have huge bits you have to scroll through.

Things like jump to links can help Google as well get you to the right place in the page from the search. That's something we've seen Google doing recently, which is you do a search and when you click on the result, in Chrome, it jumps you to the correct part of the page and it highlights the text you need in Chrome as well. Again, I tend to avoid putting things if possible behind entire tabs, because it just makes things less scannable. The last UX principle that I tend to think about is, because lots of devices we're not using a mouse, we're swiping, is that a scrolling action on a mouse as well really, is a little bit less friction than a clicking action, especially on mobile where you have to be a bit more accurate. It's easier to just as quickly scroll past something than open up tabs.

There's no real issue anymore with, if you do need to hide bits, as long as there's the caveat of, as long as you do it in a way that search engines can still read what's behind whatever you're hiding. It's certainly now I would say common best practice when we're dealing with mobile first pages.

NL: Mark, we're up against the hair, with plenty of questions coming in and thank you all so much for asking your questions. I promise I'll get around to as many as we possibly can. Mark, coming back to you with Alexis's question, is there a limitation range for the sum of keywords to be used per image? Thank you in advance.

MC: I think of it like spreading butter on your toast, in that you've only got so much, Google's only going to pay attention to so much data that you put in. If you listed, for instance, every single word in the English dictionary on an image description, that makes no sense, right? And there's going to be a certain amount of importance placed on each key word that you use. If you just start using thousands, it's going to be spread very thinly. Okay? I would focus on the things that are most descriptive. If you're using an image, I would just say, can you describe what's in that image without someone seeing it? Especially with things alt text, you just want to use language to describe the image. You don't just want to list words.

If you had a picture of a black Labrador in a sunny field, that is similar to what I would use as the alt text, I wouldn't write things like, dog calmer, black calmer Labrador. I would actually just describe it that way, because that's more helpful from an accessibility point of view. Again, caveat, if you are writing alt texts, don't write a photo of image of anything like that, screen readers do that for you.

NL: Thank you. And Alexis, I hope that helps, do come back if there's anything else you'd like to ask on top of that, same with yourself Nadeem. Come unto yourself, Rupa, who's asked, how do I deal with out of stock items?

MC: This is a good question. Getting them back in stock would be the first answer. This is a bit of a trap, because if you say on your actual product page, this item's out of stock, or if you use schema to say it's out of stock, Google will try not to rank that page if there is another page on the web with the same product that's got it in stock. And of course that makes sense, right? To users, why would you want to be sent to a page where something's out of stock when you can go somewhere where it's in stock? I have seen certain e-commerce retailers be very sneaky, they will never use the term out of stock. They will explain, they'll do things like email on availability, for instance.

Firstly that's always something I would do with out of stock. If you know you're getting it back in, give an estimated date when it's going to go back in. Offer an email to ping the customer when you've got it back in stock. If it is a product that's stocked in lots of other places, you're always going to be up against it, because if people can go somewhere in two clicks and get what they want straight away, it's unlikely they're going to wait. There are exceptions, if you're selling one of a kind or unique things that aren't available elsewhere, it doesn't matter if they're out of stock because nobody else is going to rank anyway. While it was a joke of getting it back in stock, that's really the answer. Basically, you have to accept that if you're saying it's out of stock, that Google's probably not going to want to rank you.

Or you can try and get around it by being creative with the wording and you also have to drop the product schema is what's called offer. And that describes things like if it's in stock. Now, the offer part of the product schema is not compulsory. Meaning the other thing you need to do if you want to be sneaky, is if something's out of stock, you actually don't deliver the offer schema anymore. You're describing the product, but you're just not bothering to mention the small detail as to whether it's in stock or not. I think to be honest, that's a bit hacky and I think Google will, I don't think they'll penalise people for it, but they will work out what people are doing there. It's got a limited shelf life. If that's going to be a huge amount of development work for you, I'd say it's probably money better spent elsewhere.

NL: Mark, thank you. Stanley, I'm coming to you next. We touched upon this a little bit earlier, but he's talking about discontinued products. What are your thoughts about redirecting them to their respective category pages? My impression is that they should be redirected to another page with similar content as much as possible, but it's fine to delete the page as a last resort if it isn't getting any traffic or ranking well, and there are no suitable alternative pages to redirect to.

MC: I'll share it with you after this. I actually did, I think it's quite a nice flow chart for discontinued products for SEO. Again, it's one I've tried to make generic because different sites will have different requirements. But the general flow is, yes. If a product is discontinued, my first question is, is there a close product that you'd want to send people to? And if the answer is yes, then I would keep that page, say it's discontinued and then say, you want to go over here now. Right? Keeping with that, what's likely going to happen over the months is that the search volume for that discontinued product is going to drop off and people will start searching for the new product.

When that happens, I would then actually do a 301 redirect from the old discontinued product to the new one. I set an arbitrary time, it was three months or something, that I would generally do that. Because of course you don't want your site littered over the years with hundreds of discontinued pages. Again, it doesn't really make sense. And of course you need to remove internal links to them. But that would be the situation there. The next step is, okay, we've just discontinued the product, there is no real one for one match for this product. Right? Now, this is where it gets a little bit tricky and you have to just use common sense, because this depends like, okay, is there still a matching category intent for that product or not? How many searches are we getting for that product?

If there is that matching intent, then again, if you've got searches for that page, I would still normally put a discontinued page up for a little bit, just to again, have that good user experience of, you've looked for this, we don't sell it anymore. Go look at this category. Okay? Again, after a certain amount of time, I would then 301 it to the category level, most likely. And then the final step is, okay, we've discontinued the product. We've discontinued that whole category. There's nothing relevant anymore. Then it would come down to, okay, does this URL basically have any links to it? Does it have any other websites citing it? If so, I would try and find somewhere to redirect that to, because links are the lifeblood still of ranking and you don't want to get rid of them if you can avoid it.

If it has no links or anything worthwhile, then again, the first step would be to discontinue the page, explain to the customer why it's not there. And then I would do a 410 instead of 404, which there's a small difference. Search engines treat them the same, but 404 is obviously a page not found. So you've asked, this URL doesn't exist. 410 is the service saying this page is gone, as in, it's been intentionally removed, we know, chill, don't bother coming back. It's not coming back. They're the three main, I've done a flowchart from, I said, I've worked through, again, obviously there will be exceptions for different types of businesses and different situations. That's roughly the wide following. So yes. Question and I was correct, in that there are absolutely cases where you would just cut it off and not redirect. There are cases where you do category redirects and there are cases when you'd leave it discontinued.

NL: Thank you, Mark. Stanley, I hope that's helpful. Francisco coming to you next. He says, I'm sorry Mark, a bit out of scope here, but why for some websites on Search Console, do you still get the desktop crawls, the main one instead of the mobile one, seeing as we're in a mobile first world? Not sure if you know anything about that.

MC: Yeah. Because Google hasn't got around to doing mobile first crawls everywhere yet, because the web is very big, basically is your answer. They had their original goal, I think it was March, I think it was March a year ago that they were going to have everyone mobile first crawling. A few people questioned this and Google is just kind of, yeah, we're getting around to it. The internet is really big, bear with us. Again, no reason not to be optimising for mobile first. It's just, it will change. It's not rare. I still see that on some of our sites that we look after.

NL: No problem at all. Akash coming to you next. Top three points Mark, when it comes to optimising videos for being indexed and ranked in Google, if you were to choose three things that you should do from today, what would they be?

MC: Good question. Just the basics really. So schema, video sitemaps, and think about the actual video title. The absolute basics. Again, if you are really getting into video, then you need to start looking into YouTube, then you're looking more at engagement metrics than anything else. If you're going to do a video, just please do it well. Audio is the thing that catches most out on video. It costs about a hundred, 150 pounds to get a very nice mic that will sound decent. There's nothing more painful than watching someone that is just recording off their phone five meters away, it is horrible.

NL: And so Sean has put a piece in here around Amazon has lots of currently not available pages, but they still rank for many keywords. I'm guessing Mark they've adopted one of the tactics that you mentioned earlier in terms of, just not quite updating the schema or something like that to get around this.

MC: I just wouldn't compare anything to Amazon. There's just nothing useful that's going to come from that. Amazon, is an edge case, an absolute monolith of the web. There's a million things that's going to cloud any type of analysis you do from them. From an SEO point of view, apart from the fact that everyone knows who they are, they've got this huge brand presence, they've got millions of pages, they've got millions of affiliates. I think it's useful to look at the best practice in terms of what they're doing on their pages for users, because they spend a lot of money and they have lots of systems to do that research for you, but in terms of why are they ranking, why they are not. The answer is shrug it, it's Amazon, you're never going to be able to have the same play in that environment that they're in.

I would actually try and look at people that are near your competitors and maybe why are they ranking, why are they not. I can go on for so many reasons why Amazon is different from other sites.

NL: That's fine. Thank you. Let's come to you, Nelton. What are the best practices for menu navigation? How does it impact if a menu has a hundred links in it from the drop down menus and different categories and subcategories versus perhaps five or six?

MC: That's a really interesting question. Again, you've got this mirror between user and search engine. The internal linking for SEO is really important. And internal linking is talking about how you'd link your own pages together. Now, what algorithms like page rank are describing, is essentially how important specific pages on the web are, based on how many other pages they're linked to and how many pages those pages are linked to themselves. If you created a whole new internet, just with a hundred pages and none of them are linked together, they all have the same base amount, tiny amount of page rank. And when you start linking them, that's when it starts gooping up in various places. So how this impacts you from a menu navigation point of view, is that, essentially any page that is linked to from your main menu, is by definition linked to on every page on your site where that main menu is at least which is normally all of them, right?

Any page that's linked to that main menu from a user point of view, seems very obvious, is obviously very easily discoverable. We'll leave it there as easily discoverable. From a search engine point of view, that reflects through page rank. And what you're basically saying, is that any page that's linked to on the main menu, is a very important page because it's linked to from lots of pages and those pages it's linked to from, are linked internally lots as well. You need to reflect this in your planning if you are a link architecture, which is okay, these are competitive categories or subcategories. We want lots of people going there; this is a sensible start, the kind of place that they probably want to be in your main menu.

Personally, and this is a personal preference, I am quite a fan of similar - actually to give you an example from a user point of view, from Amazon, using those mega menus, they're pretty big, right? I personally like them, from a user experience point of view, having that menu instantly, it's like a mini contextual site map, tells you they sell all these categories of products. I've got some useful information there. They do, or don't sell what I need. From a search point of view, you're getting that perfectly optimised anchor text links from important pages to all of those. If possible, I try to have that approach slightly flatter, rather than having to click to a category, then click to a sub category. And then if things get really horrible, a sub category, you really want to avoid that.

It does have a big impact on SEO. You definitely need to have what you're doing tempered by someone who, and this isn't me, someone who knows a lot about UI and user experience and has a design side. That's why we work with designers internally when we do this kind of thing, so they can help us with best practice. But yes, definitely one of the biggest things you have control of, is your internal linking.

NL: Thank you, Mark. You must be saying something all right, because we're getting so many questions, I'm struggling to get through them. But let's come to Matta next. Who said, how should he choose the best product for certain keywords to rank? Is there any process you would go through to say, okay, this is my product, and out of the 10 key words, this is the one I'm going to choose?

MC: Sorry, is the question though, the question was choose the best product for keywords or choose the best keywords for the product?

NL: Best product for the keywords.

MC: Okay. Best product for the key. That would be, I think, almost on its head there. If I had a product, I wouldn't be looking at keywords and then trying to find a product to match. I would be looking at the product and then trying to research what people are calling that. And the reason is, even if you've managed to rank for keywords, you've chosen for your product, unless they are actually really what people are searching for, you've got that mismatch of intent, which is then your conversion rate will be lower, Google will eventually work out, this isn't the right thing. If you've got the product, how do you do keyword research basically?

Firstly, you've got your internal language that you use to describe that product. Secondly, ask your customers, right? What do they call this? What is this thing? What would you search for? This is literally the first step, just before you use any tools, just ask people, ask random people if they're relevant, or ask your focus people that are your customers. If you were looking for this product, what would you search for? That will give you your seed phrases. Then you can start using things like the Google keyword planner, first step, which will give you basic things like search volume for those. Obviously if there's a clear winner in search volume, that's where I would start my exploration. You don't necessarily always want to go with the highest volume.

If you've got a budget, if you've got PPC campaigns running, I would, again, that's a really great place to get data from. Just bid on those key phrases, spend a couple of hundred quid and then see which ones actually convert. Okay, this one only has half the amount of searches, but it converts at three times the rate of the one with more searches. It might be, you want to focus on that one first, or it may be that everyone is piling on to this big search term. And this happens a lot, right? People do keyword research, they pick the one with the highest number, and then everyone optimises for that. Sometimes you just pick the second one down and you realise that you're then, that's got 80% of the searches, but you've only got 20% of the competitors, because everyone's just opted for that one with the highest search.

That can be a real good end to very quickly get sales. Then you want to start going down the line of using tools, like Answer The Public and our tool, it's down for maintenance at the moment, but it'll be up later this month, alsoasked.com. Then you can start gathering questions that people are asking around these kinds of products. That's things you want to include maybe in FAQ sections on your product page and capturing the long tail search. And then branching off from that, you've got normally a whole load of things you can produce about again, how do you use your products? How do you maintain the product, comparisons to other products? Again, coming back to, what do people need to know when they buy this? What are they considering? What's their intent?

The actual keyword selection for the product is, there's a process there, it's fairly straightforward, it shouldn't take ages, but then as you branch out into the, all the intent that surrounds products, that's again, it's more effort, but that's where you can get a lot of value from SEO. You get a lot of traffic because you're doing things other people aren't bothering, it's a way to get links. It's a way to gain trust if you're actually showing you're the person that's actually providing that value as well as just selling them the thing.

NL: Perfect. Thank you, Mark. Coming to the next one from Raul, that says, do you recommend optimising for an e-commerce site with affiliate keywords, such as best niche products or top niche products reviews 2021? In his point of view, you only sell products from your brand, but keen to see your thoughts on that.

MC: In the e-commerce space, you've obviously got the merchants who are people who own the e-commerce sites, who own the brand, the products that you're selling. And then you do have this affiliate space, which is trying to rank in Google by producing extra or additional value. They are the non-bias comparison between brands, between products, independent reviews. Personally, I don't think e-comm brands should try and do that themselves, because everyone knows that you're not unbiased, you want people to buy your products. It's a little bit silly being like, okay, here's Mark's furniture store, here's our fair comparison to our rubbish competitor and their flimsy chairs that they sell. It's just not going to work.

It should still play part of your SEO strategy though, in that, for instance, I've worked in the electronics niche selling TVs online, and apart from optimising the site, one of the things we were doing was looking at searches for people doing things like, what's the best flat screen, widescreen curve, whatever TV? And you'll see certain websites that rank, and they are these independent maybe affiliate sites that review them. And all you have to do is actually get in touch with them and just be like, we're from such and such brand, we think our products are great, of course, can you include this in your next roundup you do, or whatever it is? You can still have mentions on those pages, you can still get links from them. And actually, I think it's better for you and the consumer to see you on these independent sites.

And again, even if you get a shoddy review, you've got some good product feedback to go on. I would personally avoid, there's better things for you to focus on. I've been doing that affiliate stuff.

NL: Thank you Mark. Guys, about 10 minutes to go, please keep your questions coming in. I can see I've got two or three more rounds to get to, and I'll try and answer them all before the end of today's show. The bit I love about the series that we're doing together, Mark, is that we're getting guests from all over the world. It's not just all from sunny Norridge, they're dialing in from all over. Thank you all for getting so involved. Rupa, I'll come to you next. You asked a cracking question earlier, it looks like another good one here. He says, suppose I have to use product description content directly from the manufacturers, that the content is not unique from my website as they will be repeated across multiple websites.

Obviously Google doesn't duplicate content. What do you do for larger e-commerce sites in that case, when actually you've almost got your hands tied by, this is our content that we have to use for our products and you're reselling on their behalf?

MC: I've never seen a situation where you're not allowed to add to that. There might be a situation whereby you have to mention specific things. But if we play along and say, okay, hypothetically we have to use this boilerplate content for product description, then I'd look at the most important on page elements, things like the page title, seeing if we can describe the product slightly differently or actually look at this supplemental content strategy. The most common thing that I actually encounter with that, is just sites with thousands of products and you've just got the feed and everyone imports the descriptions, and then we're stuck with the task of, okay, well, who wants to rewrite 10,000 product descriptions. It's like tumbleweed; nobody wants to do that.

There are actually lots of tools you can use now, Copied or AI is one that can give you a good headstart in writing content descriptions. That's something I'd look at if you're trying to do it on scale. Otherwise I just triage and look at the most popular products and start there and just work my way through. I've certainly never encountered a case where you're not allowed to add additional information to the products and that's where I'd go with that. Because you're absolutely right. If you've just got an identical carbon copy page with the same images, it's not that you'll be penalised, but Google will filter the results and just pick one.

NL: Thank you. I'm going to come to Devron next, who, it looks like you've got a fan over in Sydney, Mark. She says, I can't wait to use alsoasked.com when it's back, but would you recommend adding product reviews into product pages? Is it beneficial in terms of ranking?

MC: It depends, I guess, on the content. I would say, I would always include product reviews on product pages, is the short answer. Yes. It might be that people were asking questions or not asking questions or leaving comments that, again, match up with the types of search people are doing. You've got product review, schema as well, that can help. Google has loads of really interesting paintings around trying to work out sentiment around brands and things, entities on the web. Having a painting obviously doesn't mean that they are actively using them. I just can't see any downside to doing that, again, unless the reviews are bad, in which case the problem isn't that you are including the reviews on the pages, it's that your product sucks and you need to fix it or something's wrong.

I would always do that, again, look at Amazon, it's one of the first things that people look at as well. If you run something like Hotjar on product pages, when people know you've got reviews there, the process for a lot of people is, okay, this is the correct thing that matches what I need, is the correct price. Now I scroll down to see if it's any good or not, because I'm just going to read some user reviews. Okay. I'm satisfied with that, I'll go for it.

NL: Final question for today, Mark is, around the importance of voice search. For those looking to embrace voice search, where should they start?

MC: Voice search, that's an interesting one. I had people saying that by 2020 half the search that's happening, were going to be by voice. I think voice search for me isn't quite as exciting yet as I think some people are making out. There's a lot of searches happening that are functional. Like, what is the weather tomorrow? Set my alarm and how many feet in a meter, that kind of stuff. When it comes to people doing transactional searches, the most common thing we see is them actually just using their voice to type the search query in, rather than actually using really any assistance. Last week, Google announced that their Google Duplex now is going live for supported restaurants.

I don't know if people did see this, I imagine they did, because it was viral. In May, 2019, we saw Google demonstrate their Duplex technology, which is their thing they use with Google Assistant. I don't know if you saw this Nathan, where they were trying to book a hairdresser's appointment and then the Google AI phones up and just has an actual conversation with the shop. Did you see this?

NL: No. [crosstalk 00:58:58].

MC: Okay, I'll have to share this with you. [crosstalk] This is really big, right? Google had this tech called Duplex, it's like this assistant and you say, go book me two seats in this restaurant and it will, in the background, phone up and have this conversation saying, I'd like to have a reservation for two people at 7:00 PM. And then they might go, we haven't got 7:00 PM and the AI's okay, well, what time do you have? And they're like, we've got 8:30 and they're like, that's fine. It has a full on conversation, right? That tech combined with the latest announcement, a couple of weeks ago was Google's releasing this set of systems that they're broadly calling MUM, which means that when you do a complex query now and by a complex query, I mean maybe the example they gave is, if say you had hiked up a Mount Fuji, and then you were going to do a different mountain and you want to know the difference, usually you'd have to do several searches.

This technology predicts what the several bits of information you need are, and it doesn't just give you a webpage, it gets the information from the websites and then using texts similar to GPT-3, it just writes you an answer. It basically provides you a custom webpage on the fly, that's got exactly what you need on, right? That means that, if you asked a mountaineering expert, what's the difference between these mountains? They would know, okay, well, I need to tell them about the shoes and the rainy season, which is a different time of year. That's the kind of thing it's doing and it would just tell you. We've got those two massive bits of technology, which are just inching over the line. I think in a few years, once they are over that line, that's when voice search is going to become exciting, right? Because then you can actually conversationally say, it's 35 degrees outside, what do I need to go running?

And Google will be like, okay, hat, some protection, camelback, these types of shoes, that particularly airy kind of thing. That's when voice search becomes exciting. For me, the tech is still pretty basic, when you watch people use voice search, it's frustrating. They're shouting at the thing, it's getting words wrong. It's just not the slick thing you see on Star Trek. Right? There's nothing particularly you need to do for voice search, except, I'm going to say it again, schema, schema, schema, schema. That's the key to all of this stuff, is schema, right? The key to Google being able to use this tech, the key to us being able to use it as consumers, isn't search engines, reading web pages, and heuristically working out what's going on. It's about people providing structured data that they can put into a graph of edges and nodes and it understands what's right to what, and then the AI can move around that.

The short answer is, schema, but the backup answer is that in my opinion, there's not a lot that's going on. You're not going to double your sales because you did a voice thing. But that in my opinion is coming and web search will be in the bin in several years time.

NL: Sruthi, I will just say, if you look at the last recording and you look at today's recording, hopefully that will answer your questions about where to start with SEO for newly created e-commerce businesses. I'm conscious that we're coming towards the end of today's session and I just want to finish by suggesting that everyone that's enjoyed today, please do like, share, engage with the post, tell others, get others along. We had a hundred to 150 people along today. We're already looking at about 200 to 250 for the next one, which is on the 30th of June, that's nine till 10, where we'll be taking, or tackling another big topic within the world of e-commerce and SEO. But finally, it's just left for me to have a massive or to say a massive thank you to Mark, who once again has given up his time this morning to share his wisdom.

There seems to be nothing that this bloke doesn't know when it comes to SEO and e-commerce. Please join us for the next one. Mark, just very quickly, if people want to know more about you or Candour, how can they find out more?

MC: I'm pretty active on LinkedIn, just Mark Williams-Cook. I post tips everyday on there. I run a podcast every week called Search with Candour. So if you just search for Search with Candour, you'll be able to find it. That's 20, 30 minutes every week talking about PPC and SEO. If you're interested in working with us, just search for Candour Agency, hopefully it will be top, and you can see what we do in the realms of search, but always happy to talk to anyone about SEO or search in general.

NL: Mark, as always thank you for your generosity and your time. Great to catch up. I hope you've all enjoyed it today. Thank you so much for joining us. We look forward to seeing you again on the 30th, and until then best of luck and keep in touch. Thank you.

MC: Hope very much you enjoyed that recording. As you heard there, you can catch those sessions live on LinkedIn, so you can find me Mark Williams-Cook on LinkedIn. We'll be back next week on Monday the 5th of July with one of our usual episodes, but we'll continue to bring these recordings every few weeks as they do happen. As long as there isn't anything super interesting that we need to cover in the SEO or PPC search space. Until then I hope you have a brilliant week and I hope you'll tune in for the next episode.