Optimizing Categories Pages in E-Commerce SEO Processes

Welcome to a new edition of Crawling Mondays. This is a very special edition and I know that I say this a lot, but this in particular is an edition that I'm very excited about because it's one of my favorite topics: categories and facets optimization in SEO. I do a lot of SEO consulting for ecommerce websites and I am sure that many of you have been working in ecommerce too. and know how critical this is, how overlooked this is, I'm afraid too. So today we'll have the amazing opportunity to speak to a couple of people who have a lot of insights, who have a lot of experience in these topics, and we're going to go through the major areas of this. Before starting the conversation, I want though to thank our today's sponsor, Content King. In case you don't know Content King, they are the SEO alerting tool. They will let you know whenever Anything is changed, if something goes wrong, they are pretty much a real-time SEO crawler. Thank you very much, ContentKing, and take a look at them in the link below. And let's start the conversation already. My guests today are 2 amazing women in SEO who are, as I mentioned before, focused in this area and have tons to share. On one hand, I have Tori Gray, CEO and Principal Consultant of The Gray Dot Company. Hello, Tori, how are you?

I'm great. Thank you so much for having me.

No, thank you for coming. On the other hand, I have Lydia Infantes Jr., International SEO Manager at BigCommerce. Hello, Lydia. How are you?

Hello. I'm good. I'm in the BigCommerce office in London today.

Amazing. Amazing. So let's start a little bit of conversation, right? I have seen also many, many, many times, many websites shoot themselves in the foot by paying only attention, only linking, only indexing. major first-level category and overlooking completely the rest. So I would like to start by asking you, from your experience, why optimizing categories facets is so critical for ecommerce?

It's critical because it allows you to show your content for very specific bottom-of-the-funnel queries. They're looking, for example, for white dresses or Chelsea boots, right? You're mixing 2 concepts into a bit of a more long tail, if you allow me that inaccuracy. And it allows you like a really good quick win piece of traffic. My very first experience with fastest navigation and categories wasn't in SEO, but it was in PPC. I was working in an international ecommerce site for personalized clothing. So optimizing the landing pages that I was showing in the right keywords made all of the difference in terms of conversions and in terms of the PPC price that I was paying. So I was working with developers trying to find the right intersection between the types of attributes that the products had that we wanted to index and the types that we didn't want to index. So there were elements to what we were selling that were very B2C-oriented and there were elements of what we were selling that were very B2B-oriented. Where there was volume in terms of organic traffic was on the B2C-oriented type of filters. So instead of talking about the number of grams per square meter that the t-shirt has, we're going to be talking about fit, we're going to be talking about color, we're going to be talking about brands, right? So we would have different types of attributes. We chose ones that had the biggest volume and we'd make certain combinations of categories indexable, give them their own page, give them their own URL. And it allowed us to reduce our PPC budget by a lot and increase our conversions. So it is not just relevant for SEOs, it's relevant all over.

It's to connect better with the user search experience at the end of the day. Yes, and the user search journey because many of those queries for which subcategories or facets we'll end up ranking for are very commercially driven queries that still have non-trivial search volume in many cases and will allow users to have, let's say, a diversity of options rather than trying to push them towards a particular product, which sometimes doesn't make sense, especially if you want to really fulfill the intent of the user that needs and wants to choose, right? And there are many different products to fulfill their need, right?

Yeah, it's a lot about user choice. And from my perspective, honestly, it's the bread and butter of ecommerce. I mean, obviously, product pages are critical as well. But if you don't get category pages right, you're minimizing Google's ability to crawl to and pass equity to the product pages. So you're literally limiting your success by not setting up your categories and your facets well, because you're not going to flow PageRank to them, they're not going to have the ranking potential that they would have otherwise. Plus that user choice. It also is tied to your brand, frankly, like what categories, what facets are you choosing? And how well does that emphasize what's meaningful to your brand? You can help tell a story about what's important to you and why they should purchase from you instead of somewhere else. Like that's user communication. It's really critical.

What I have seen in the past is that many websites and many decision makers at the end of the day, very afraid to start indexing, for example, certain type of facets or filterings or start linking to certain pages because for them it's like, what that will do to my crawl budget? Or will they generate content duplication issues? Or I know that I should be doing something with all of these listings from different subcategories, but I don't because I don't want to add any content to these pages, will hurt my conversion. So what is important at the end of the day is to have well-set criteria and well-set goals also in place to make sure that those changes that are implemented are aligned to an actual supply and demand that at the end will generate profitability, right?

Yes.

But before moving into that, let's go through a few of the mistakes. I want to ask you regarding those types of scenarios or just those types of actions that people end up doing because of this perception of subcategories and facets. So for example, in my case, what I have seen are this rule of thumb. So after second-level category, let's canonicalize all subcategories to parent categories. So none of this will be indexable because all of these facets will be canonicalized to the parent categories. All colors, green, yellow, whatever, no, we're too scared to index this because we're unsure that the content is good enough or there's enough content to index individually. Then I have also seen A lot of scenarios of complete navigations linking internally to non-canonical URLs, to URLs with UTM tags, for example. What are the main mistakes that you see websites doing again and again?

I think that the things that you have to nail when you're doing navigation is going to be pagination, hreflang, and canonicals. Those are the technical aspects of it. To be able to make the right technical decisions, there's another step before that, right? And that is market research. You need to understand what you have, what your users want, and what is the user journey to get there. So you can work with your data team to create a matrix of all of your categories and subcategories in a big matrix of all the variations in which all of this can be searched and then try to extract keyword volume. That way, you're going to get a general idea of demand around your categories and around those specific search volumes. You can have a chat if you want to refine that data with qualitative data with your sales teams, with your support team. When I was working in that international ecommerce that I was mentioning earlier, that was so, so, so key because they were like my eyes and ears of what the clients actually wanted and the ways that the clients actually spoke about our product, which wasn't necessarily showing up in search volume. And once you have that, you need to understand how to implement it. You need to make the choices of what is worth making canonical, what is worth giving it its own space, what you can scale and what you cannot scale, what can you automate, and where are you deciding to leave money on the table? Because it's not going to be perfect. It's never going to be perfect. You're never going to be able to account for every niche case and every little search. But at some point you need to say, this is where I stop optimizing. This is enough revenue. Let's talk pagination. Let's talk hreflang. Let's talk canonicals and get those implemented correctly in a way that's going to reduce friction because complexity and deciding what's good enough instead of aiming for perfect. It's perfect. It's not going to happen.

100%.

I think it's so critical to establish criteria based on the demand and the supply that you have and the goal towards the revenue and profits that you're looking to achieve with the different products that you showcase across all of this. different types of categories. At the end of the day, like configuration of these pages should pretty much reflect that, right? The problem is when you start optimizing without having the north star, what do you want to achieve on one hand? And then on the other hand, the input regarding user demand, audience demand, potential customer demand, and what the customer actually wants.

Yeah, I echo that. I mean, pagination issues are common, especially with the canonical issue. Like they always want to canonicalize back to the main page and like, Well, do you want page 2 products to rank or not? I mean, what do you want? The variation that I often see in terms of the keyword research is especially like manufacturing brands wanna call things by technical terms and not the language their user uses. And so there's a big disconnect in terms of what the opportunity is for to find people using their language instead of yours and where you are at. A big one that I see is just frankly creating categories and facets that overlap with one another. So let's say we're talking about tables, and then they have a category, a subcategory of round tables, but then they also have a filter of round tables. And then just like that at scale gets really, really messy. So it does tend to go one end of the spectrum or the other. Like you're afraid to index anything and use the tools that are in the toolbox that we have to do this, to control this well. Or it's just Google has access to everything and everything's a duplicate and they're all fighting with one another and Google doesn't know what the heck to do with it. So that You have 1 or 3 products in a category page. It's a sub-sub-sub-sub-subcategory. And it's just like, why are we doing this? This is pointless. This is not helpful.

Thank you very much for bringing that up. It's like the extremes, right? I think that we can leverage tools like Semrush or Ahrefs that tend to do this at scale for any query, right? Not only for those that we specifically track to show, for example, when there are a lot of products URLs. Ranking and continuously being replaced with one and the other for queries that should be targeted by subcategories or facets, right? And then you say, it's like, why is Google ranking product pages? And there is a blend of many product pages because Google don't know which one to rank. All of them are somehow relevant for this query. But of course, since the facet is noindexed or canonicalized to the parent category, Google won't index the actual facets. Then what you then realize is that if you go and actually search for that query, an example, red cups, and then you see a lot of individual red cups products pages ranking for this query, even if they end up ranking somewhat well in SERPs, it's confusing for the user which one they should click. Will they click particular specific cup pages, or they will go to that other player that is maybe not— it's not ranking in the second or the first positions, it's in the third, but they can see it's a category page that will offer them variety and diversity of products where they will be able to choose. I will click on the third position, right? So it is also improving your click-through rate by better matching the customer intent or the user intent. Let's move towards the building blocks. To avoid all of these issues and all of these challenges, what would you say are the fundamental steps to optimize categories, subcategories, facet pages? Which are those criteria that you use again and again, or the steps that you follow again and again?

For me, it always starts with the data about what people are looking for. So that can be your keyword data, which is a great dataset about what we know people are literally looking for. But it also is your sales teams or your customer service teams, and what do they know about what your users want, and what's meaningful. You have to know what people are looking for that you can provide so that you can architect a plan for how do you want to organize your site? Are you gonna have 17 red mugs? Then okay, then it makes sense to make a red mug category page. If you have one red mug, okay, then maybe you optimize the product page for that. But you have to know what do you have and you have to know what are people looking for and what's important to them in the way that aligns to your brand. After that can come all the, the architecting, the planning and which page and the on-page optimization. But you have to start with what's the need and how do you fulfill that?

Off the back of that, I'd say choosing the sources of your market research and your market insights is really interesting as well. You can have a chat with your suppliers. You can see what the suppliers demand. You have to choose your sources for market research and the market insights that you're going to derive. And search volume is only one of the sources that you can use. You can use your internal search data. Honestly, anything. This example about the red cups has made me think of websites such as Redbubble that have personalized products for specific categories like nerdy pillows or gifts for nerds, gifts for teachers. But also, I personally buy a ton of Sailor Moon merch from Redbubble. So I know that that's something that's not necessarily a category of yours within the product, but it's a category of how your people are going to search for your product. So it's not just within your inventory. But within the user journey and what the demand is like. So yeah, it might be a white personalized mug, but it's definitely a Sailor Moon mug in the way that I'm going to serve it.

I love that you both mentioned the demand and user preferences and user behavior. At the end of the day, that should be a main driver. There are 2 main drivers, the supply and the demand, right? This can change a lot and we need to be flexible and susceptible to analyze how dynamic is the inventory on each site to be able to set rules that will play well with crawlability, will give also a good user experience.

I was just gonna say, for some ecommerce, inventory is gonna be quite stable, but some others are going to have a supplier going bust, for example, right? Which is something that throughout the pandemic, many ecommerce sites have had to face. for the very first time or different regulations are going to change how much you can get of a certain product. It is very ever-changing and how static and how dynamic it's going to be changes from industry to industry, from website to website. And it is truly a decision that there is no one-size-fits-all solution, right?

Yeah.

It is extremely personalized to your very own business. To your users' search journey, your market, your competitors, your suppliers. It really goes beyond search volume and it goes beyond SEO. It is a business decision.

Yeah. How replaceable is the product or interchangeable is it? Do you want a wishlist? Are you willing to wait around for this thing or is it just any other red mug because you just need a red mug and you don't care?

Yes, exactly. So is it a PS5 or is it a black Chelsea boots from Timberland, right? I don't really mind if I get one model or the other. But a PS5? I wouldn't use a Switch instead, right?

Moving things towards a little bit of more the content side of things. Featuring more content in categories or subcategories. We sometimes analyze many of the SERPs. We assess the pages that are being ranked and we literally see how the pages that are ranking better are featuring more products, have more products. Instead of listing 12 products per page, they have 60. So of course, more products better fulfill the need of the user. And then all of a sudden, we realized that these pages also blend very well insights and FAQs that are actually useful for the user at that stage of the journey when making the decision of what product to choose. This is a content-rich page, but it's also a commercial page. And you see how the content is actually supporting the journey as a conversion rather than harming the conversion, which is what many decision makers sometimes are afraid to do. And this is how we end up having— I'm pretty sure that you have seen this many times— of the category pages with a how-to guide at the bottom, but very well hidden at the bottom with small font size because of course they want to add more content. But They don't want the user necessarily to see the content because they are too afraid to harm the conversion. But then at the end of the day, featuring this content that will be disconnected from the user intent can end up even harming the ranking because Google won't know what to rank for this type of pages on one hand. And on the other hand, this content won't be actually useful. This content will very likely end up also, yes, harming conversions and even the page experience and won't serve for anything, right? So it's how you blend things, it's how you integrate things. How UX plays a very important role here is critical to work alongside.

I have an example of that actually. Like I was doing some shopping for the holidays for my husband and I'm going to a new, I'm looking at like dress shirts and it's a new brand. They look lovely. I wanna go, you know, but do I get the slim fit or do I get the super slim fit? Like, I don't know your brand. I don't know the difference between those 2 things. There is strategic content that can help me make a good decision about what's the right fit for me. I know the size, but the fit will vary. Right. And, and so there's contextual information and buying decisions I'm literally blocked from making because you are scared to show me strategic content. Like strategic content isn't just random blog posts about the 7 things you should buy. It should help the buyer journey. And if you're not thinking through that, you're, you're making the wrong strategic content.

Yeah. There was a gap in the consideration. Part of the funnel for you. And that makes total sense. I would say to what we were saying, Aleyda, about these very rich pages in content, more is not always better. Featuring 60 products versus 12 can lead to choice paralysis. If I look at like a very, very large page of many products and it is a high-stakes decision, I can go like, oh my God, I don't even know, click away, right? Go to someone who's making it easier. But also, what's really important to understand is that content costs money. Either because the actual money that you're paying a freelancer or the hours of your team. So you need to be very mindful of what the user intent is when they're arriving to your pages. What are the needs? Why are you driving them away? You can do some qualitative analysis or use your data. And don't be afraid to test and choose what pages are going to deserve content and which ones don't necessarily do because you're going to harm your ROI. You can't invest the same in every page if not every page is giving you back the same and not every category page is going to need the same amount of handholding when it comes to the content. So it really comes down to think of your user and not search engines. Know your market, know your product, know your brand, and understand what the user is trying to do.

I think that you have brought up something fundamental, right? That is about not follow rule of thumbs, but actually testing. And we have so many more tools nowadays to do not only UX, A/B test to see how it did. Oh, featuring twice as many products is actually going to harm user experience and users will end up clicking less and buying less because of that, right? And you can test that. Also, we can now also test this type of things from an SEO standpoint too. Featuring content in my top-level categories, let's test if that end up actually improving the traffic of these categories versus these others where we haven't added or featured any content. We also need to take into consideration that many times it's not about doing the things or not, but how they are done, right? There are so many harmful ways to display more products that will end up not only making the pages slower, but also the experience poorer because of the type of design. That also quite a few, many other ways to do it that won't have the same effect. Many category pages that I have seen, they don't even look like category pages, right? They are so beautiful. They are well blended. They are amazing. So it's about testing, evaluating, assessing, and not being scared to run pilot projects, to run A/B tests from different standpoints to evaluate what makes sense in your particular case, right? To start wrapping up, I would like to ask you, For a last tip, something that we haven't yet covered and you would like to mention because it's something important that you think that everybody should take into consideration when optimizing categories, subcategories, or classes.

I would say embrace that it's not going to be perfect. So try with done, right? Have an incremental approach to your categorization and don't be afraid to test. Because you're going to lose more money by not trying anything than you are by trying something and rolling it back.

100%.

Establishing clear criteria also to implement things and knowing like, okay, you have done your work identifying search volume, search behavior, trends, seasonalities across different business units, categories, types of products, product lines, etc. And understanding What is that threshold that you will start with that shouldn't be perfect at the beginning, that you will refine along the way, but should be good enough at the start to start testing?

Okay.

I will index beyond that. I will noindex after that. I will configure this in this way after that. And then continue refining after that. I think that is also important to acknowledge that it should be like a continuous testing and refinement process because we will tend to manage or work with Hundreds or thousands of categories, sometimes hundreds of categories. And this is something that we shouldn't expect to do in one-time implementation, right? It needs a lot of refinement.

The other one I'd add is to think about link building too, especially in terms of your category pages. It's such a common miss to spend time building links to products that go away in 6 months to a year. Sure, link to your product pages if it's your money product and it's going to stay around for years and you know it'll always be in stock. But if not, focus on your category pages, because they'll always be there. And they'll always be passing equity to your product pages. It's just a more sustainable long-term plan for really a high-lift, high-effort activity.

That is a great call. Also, product category pages will give much more diversity and options and alternatives to users than product pages. But you see many times that many of the pages are attracting more backlinks. Guides, blog posts, they always will tend to link to the actual final product that they want to promote and endorse rather than category pages that will include or feature those products. And unfortunately, yes, when you do a link audit, a backlink audit of many ecommerce websites, you will see a ton of these old product URLs triggering 410s, triggering 404s after a while. Many of them being redirected, but also through many redirect chains also after a while. So why not try it instead? And I know that sometimes it won't make sense, sometimes it won't be natural, but whenever it is, try to endorse and link to those category pages or first-level facets. So thank you, thank you both very much for so many insights, for so many ideas and inputs. I really appreciate all of the experience that you have shared in today's Rolling Mondays. I hope you have seen Lydia and Tori are amazing. Start following them over Twitter in case you haven't. You have their handles here. Thank you very much, Lydia and Tori. You have been wonderful today. And thank you also very much for watching today's episode. If you haven't yet, remember to like the episode, subscribe to the channel so you don't miss any videos out, and very looking forward to see you again in the next episode.

Bye!

Scrolling Mondays. Bye-bye.