2

When it comes to SEO/SERPs, does Google read tags only in their entirety?

For example, let's say I have the following URL:

www.example.com/posters/red-posters/

Here, we see a Parent Category and a Child Category; 'Posters' and 'Red Posters'.

At a glance, it would be fair to say that Google may Crawl both pages and assume they are about 'Posters'. Consequently, increasing the chances of keyword cannibalization. The reason being obvious. 'Posters' appears in both pages.

Say I then insert <h1>Posters</h1> into the 'Posters' Category Page and <h1>Red Posters</h1> into the Child Category Page, would this tell search engines that the Parent Category is about 'Posters' whilst the Child Category is about 'Red Posters' and nothing else? Or would Google still extract 'Posters' from 'Red Posters' and factor this into their SERP calculations?

I understand that SEO/SERP is more complex than this and that tags are just one part of the Algorithm, so can we just assume that both Pages are equal in all other areas, for the purpose of this question.

  • Search is not about keywords. It is about semantic meaning. Each / represents a semantic cluster to be understood individually. Having a word appear more than once is not an issue as long as you are natural and not abusive. The whole notion of keyword cannibalisation is, in my understanding, a myth more than fact. Do not over think this. You are fine. Do what is right for your users and makes sense to you. Cheers!! – closetnoc Jan 25 '18 at 2:59
  • Thank you for your response (and time across my other questions). I guess in my quest to develop a 'perfect' strategy, I am diluting my own approach by taking on board, information from a multitude of angles. – Craig Jan 25 '18 at 3:04
  • What you will find in agreement here from many is that most online SEO advice is BS. The fact is that the whole page and site is taken into account. No keyword matching, but semantic understanding of content. The more naturally language is applied the better. The more sentence like any signal is the better where it makes sense. It is not about keywords but what the URL says the page is about. The clearer the better. As for the h1 tag, think of it as a clearer title. It is used for semantic understanding of the content even though searches are not made directly against it. Cheers!! – closetnoc Jan 25 '18 at 3:23
  • For what it is worth we all enjoy helping people. The plain truth of the matter is that I find webmasters to be the nicest people I have had the pleasure to meet. That is a fact. I am sure we all feel this way. That why I am here. I am glad to spend time helping you! I do this for my own pleasure. To do what I can helps me. Even when I do not have an answer, I try and guide the process of understanding where I can. I pass on what I can hoping that in the process I am building a better world. I may not be, but I try. ;-) – closetnoc Jan 25 '18 at 3:56
3

I will try and explain a few things for your understanding. Hopefully I will clarify how links and header tags are handled by search engines and how they relate.

When a search engine sees a link, it follows it. When a page is fetched, all links found within the page are entered into the link index. Each link has at least a from (source) page and a to (target) page along with the alt text. Also consider that analysis of the link is also stored including where on the page the link is found (template versus content), where in content, analysis of the link, analysis of the content block or portions of the content block if found in content, etc. If only one end of the link is confirmed, the source page, then the link is a dangling link. Once the target page of the link is fetched, then both ends of the link are confirmed and the link is no longer a dangling link. It is possible that the target page does not resolve and returns a 404 or a 410 error making the link a broken link. In this case, the link remains in the index and is marked as broken and likely the error code is stored too allowing a 404 to be retried.

Historically, a search engine who fetches www.example.com/posters/red-posters/ will not fetch www.example.com/posters/. Why? Because www.example.com/posters/ may not have a page. It could simply be a folder within the path. For www.example.com/posters/ to be fetched, there must be a link to that page. WordPress is good at creating these pages and links to these pages and making them valuable. However, not all sites or CMS do this.

A few other considerations I will throw in for completeness. Where the links are found within the hierarchy of the navigation and other pages. Links found on the home page are more important than links found one click away which are more important that links found two clicks away. Links are evaluated both between sites and on sites using a trust network model where hops (clicks) and values between hops are taken into account. The second consideration is SERP performance. A SERP link can perform better than another which adds value. Sometimes this does not appear to make sense except that the title tag and description meta-tag used in the SERP link are more compelling. This changes over time as search changes. Keep these in mind.

Having said that, the link is evaluated however it is taken as a whole as a link. For example, for www.example.com/posters/red-posters/ will not insert www.example.com/posters/ into the link index without a link. However, the semantic evaluation of the link will break the link into elements. For example, the domain name, the path, any file name, etc. A domain name has more value, though slight, than the path which has more value than the file name and so forth. Think of this in terms of reading from left to right. The path will be broken into semantic clusters using the / character. Each semantic cluster from left to right will also have more value. This means that /posters/ has more value than /red-posters/. Again, this is only a slight gain in value. Each element of the URL and semantic cluster from the path will be evaluated for meaning. For a path, it is natural that the left most cluster is topically broader than the one to its right which is topically broader than the cluster to its right. From left to right, each cluster becomes more specific topically. This is how people think and also how people enter search queries. I say this so that you understand how links are evaluated for meaning which is different from how a link is followed and stored as a link.

Next, please understand that the headline read order remains. This includes the link to the page, the page title tag, the h1 tag, and other elements are evaluated to understand the page topic. While there is more to the headline read order, think primarily in terms of these three elements working together in the order I have given them. This means that the semantic value of the links to a page have more value than the pages title tag and h1 tag, etc. assuming that the link values are strong. This is why my advice to create in content links that are full sentences or nearly full sentences is important. Any link text of one or two words has less value than ones that offer a subject, predicate, and object that allows better semantic understanding. Strong links along with strong title tags along with strong h1 header tags offer more value that most people realize. Please keep in mind that the SEPRs will match content and not header tags. However, header tags are required for semantic value in analysis and remain important in that respect. This leads to the notion that header tags are not important for search, however, they are in that they give semantic value to the pages topical scores. They are still important. That said, semantic values of any link to a page along with the semantic value of a title tag and header tag should be complete and complimentary.

In your case, the appropriate semantic cluster found within the path equaling the h1 tag, and assuming also the title tag, even though not fully semantic with subject, predicate, and object, should compliment each other just fine and give a good understanding of what the page is offering. For e-commerce sites it is not expected that all elements be fully semantic so your example makes sense. Clear as mud?

  • 1
    That should be a Search Engine Land post! Also I'm curious, in cases where you have a CMS that can either include or exclude Category from post URL's, what happens? In other words, example.com/category/post can be displayed as example.com/post. But the category is still there, in the architecture, just not in the displayed URL. Would search engines slightly devalue the /post just because the category is included in the string? I know it doesn't matter that much in the grand scheme, but your answer got me wondering! – Henry Visotski Jan 25 '18 at 17:54
  • @HenryVisotski Thanks for the compliment! I do not study CMSs really. Most do include a category page, but I assume some do not. I wrote my own CMS of sorts where it is OO (object oriented) and uses resources and content blocks (snippets) and HTML elements as reusable objects and gets very complex very quickly. I will be rewriting it to be closer to a traditional CMS. Would search engines slightly devalue the /post just because the category is included in the string? Assuming there is no page for the category, I believe that post does not suffer due to tradition. Cheers!! – closetnoc Jan 25 '18 at 18:09
  • Makes sense to me. Thanks for the reply @closetnoc ! – Henry Visotski Jan 25 '18 at 18:56
  • Thank you for your reply @closetnoc I am using the WordPress CMS. Not sure if this would influence your answer. As a whole, I am just not 100% sure how search engines can differentiate between 'Posters' and 'Red Posters', since both pages have the same Keyword; 'Posters'. Would I have it right that both pages would be indexed for 'Posters'. Then as performance history is generated, 'Posters' would perform well for the more generic search queries with 'Red Posters' performing well for the more specific? – Craig Jan 25 '18 at 21:19
  • 1
    @Craig Links are often the key that makes up for bad SEO. Yeah. I gotcha! I keep coming across pages that completely suck that perform well. Some of it is Googles fault. Lets not sugar coat it. But most of the time it is the sites link profile that makes a site perform well. No getting around that! Cheers!! – closetnoc Jan 26 '18 at 2:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.