The Second Pillar of SEO: Relevance
In January I wrote about my Three Pillars approach to SEO, in which I outline what I see as the core aspects that make up search engine optimisation: Technology, Relevance, and Authority. Then in February I wrote about the Technology pillar, explaining some of the aspects that go in to auditing a website for technical SEO issues.
In this post I want to continue with the second pillar of SEO, which is Relevance. Where technical SEO is about ensuring web crawlers find the right pages and don’t spend too much time crawling URLs that don’t add value, the Relevance pillar of SEO is about ensuring the search engines’ index process can make sense of what the crawler finds, and assign proper relevance values to your content.
Much of the Relevance aspect of SEO concerns itself with what we’d call classic on-page optimisation. This is where title tags, meta descriptions, and content come in to play. But these are just part of the whole picture; we need to look at all aspects of a webpage that can help improve its relevance values for whatever topic it is focused on.
Beyond the basics of on-page optimisation, relevance also concerns itself with other signals that help boost the topical significance of the webpage in question. Keep in mind that Google ranks webpages – not websites – which means you need to make sure you have topically relevant pages for every keyword theme you want to target.
I’ll skip the basics of keyword research and on-page optimisation, as these topics have been written to death a thousand times over. And increasingly we see insightful content about advanced on-page SEO like this excellent post from Cyrus Shepard – definitely worth a read, as well as the other articles he refers to.
Instead I want to elaborate on some more advanced aspects of relevance for SEO which you may not realise are actually part of the relevance pillar. Let’s start with the big one: structured data.
For most SEOs, implementing structured data on a website is part of the technical aspects of search engine optimisation. This is where my opinion differs somewhat from the mainstream, as I see semantic markup as a relevance factor.
Yes, structured data is about adding code to your website, but not everything that’s about coding is about technical SEO. We need to look beyond the implementation of a SEO factor, and understand why we implement structured data.
As I said before, technical SEO is about crawl efficiency, and structured data does not impact on crawl efficiency. Instead the purpose of semantic markup is to make it easier for search engines to understand the content on your webpage. With structured data you can explicitly tell a search engine what it is looking at, and the relevant attributes belonging to that piece of content.
If you mark up your website’s product pages, you can use structured data to tell search engines all they need to know about the product; the name, brand, manufacturer, description, price, stock levels, special offers, etcetera. Similarly, with structured data you can mark up your blog posts to give additional information about who wrote it, when it was published, what the topic is, other sources it references, and so on.
This sort of semantic markup makes it much easier for search engines to understand your content. Without structured data it’s likely search engines still will be able to come to grips with your webpages, but there might be some guesswork involved and perhaps a few incorrect conclusions about what exactly you’re selling or writing about. Structured data takes away that guesswork, and makes it very clear to search engines what your content is, and – by extension – what search terms it would be relevant for.
And that means that structured data is essentially about making your content more relevant. Which means it belongs firmly in the relevance pillar of SEO.
This is also why I believe the structured data testing tool is such an important part of your toolset. What this tool does, in essence, is give you a glimpse in to Google’s indexer process and see how it treats your webpages. It also provides some useful information beyond the obvious of just testing how accurate your semantic markup is.
Similar systems like ‘Fetch as Google‘ in Webmaster Tools also provide similar insights, as does the PageSpeed Insights tool, although these are mostly applicable to the technical SEO pillar as they concern themselves with crawl efficiency. There is, however, a fourth tool Google has made available to webmasters which also applies to relevance more than anything else, and that’s the Mobile Friendly Test.
I think that for many SEOs, mobile usability would also fit under the technical SEO umbrella. Again, its implementation is a fairly technical endeavour, so it would be obvious to conclude it’s a part of technical SEO.
But here too we need to look beyond the superficial, and analyse the purpose of mobile usability. Moreover, we need to speculate how Google would factor this in to its ranking algorithms.
When a webpage has a bad mobile user experience, showing it on mobile search results means Google would also provide a sub-optimal experience to its users. If Google were to send someone that is using a smartphone to a website with a bad mobile UX, that user would simply come back to Google and try a different result. The page with the bad mobile UX would thus be seen as a less relevant result.
This is how I believe Google has integrated mobile usability in to its search algorithms. Mobile optimised webpages provide a more relevant result in the context of a search performed on a mobile device. And, by contrast, unoptimised pages provide a less relevant result in the same context.
I believe Google assigns a mobile relevance value to every webpage, which depends on how mobile friendly the page in question is. Webpages with a high mobile relevance score will, as a result, rank higher in mobile SERPs than webpages with low mobile relevance scores.
I may be entirely off the mark with this interpretation. After all, Google could also assign an authority/trust score to mobile friendly webpages. But to me that seems a mismatch – Google is trying to offer the best possible results within their users’ context of performing a search, and for me that means mobile usability is an issue of contextual relevance rather than an authority or trust factor.
So that is why I see mobile usability as a relevance factor rather than a technical issue or an aspect of trust and authority. Mobile friendly webpages are simply more relevant for mobile searches, hence why Google will, in my opinion, assign a higher relevance score to webpages that are optimised for mobile viewing.
Lastly I want to talk about internal linking. For me, this is one of those SEO aspects that crosses over all three pillars and positively impacts all elements of optimisation. For me, internal linking from the body content of a webpage to another page on your site serves three purposes:
- Provide alternative crawl paths
- Spread link value
- Improve relevance scores
It’s the third purpose I want to focus on: using internal links to improve relevance scores for your webpages.
First we need to talk about how links fit in to the three pillars model. For me, links in general serve primarily two of the three pillars of SEO. Authority is the big one of course, as it is the value of an inbound link that enhances a webpage’s authority (or, as some call it call it, the page’s trust) and thus improve its rankings. But links also serve the Relevance pillar, through the anchor text of the link.
We all know that ideally we want links with keyword-rich anchor texts (though not too many, as they can trigger a penalty). We intuitively understand that a link with ‘click here’ as the anchor text is less valuable than a link with ‘buy [keyword] here’ as the anchor text. We’ll still take the former anchor text if we must, but we’d prefer the latter.
This is because the anchor text of a link carries relevance value, in addition to the link’s inherent authority value. You can get a webpage to rank for a given search term, even if that search term is entirely absent from the page in question, just by pointing enough links to it with that keyword in the anchor text. Known as ‘Google-bombing‘, this aspect of Google’s algorithms has resulted in many embarrassments over the years.
Google-bombing is so effective because it exploits an inherent part of the way Google understands the web. As a link-based search engine, Google needs links to make sense of the web and understand which webpages deserve to rank the highest for a given query. This goes beyond simply measuring the authority that a link carries, and looks at the relevance attributed to a webpage from the anchor texts of the links that point to it.
So links have a relevance value, expressed through the anchor text. Once we know this, we see how internal links can contribute to the relevance values of your website’s pages. The anchor texts of your internal links add relevance values to the pages you link to. This applies to both your main navigation links as well as to links from within your body text.
Because these are links you control, you have complete governance over the anchor texts you use. That means you can capitalise on the value of your internal links by optimising your anchors to add relevance to your webpages for the right topics.
This is why you see so many websites link from their blog posts to their product or service pages, with exact match anchor texts. Even when the blog post itself accrues little to no authority, the internal link still carries relevance value with an optimised anchor text.
When done in excess it’s seen as a spam tactic, but applied heedfully your internal links can help your key webpages become more relevant for their topical focus, which will make it easier for you to achieve higher rankings in SERPs.
Trust and Authority
In summary, relevance in SEO is about everything that can help boost the signals Google uses to understand the topic of your webpage. This goes beyond the classic aspects of on-page optimisation; you need to look at matching your page to searchers’ intent as well as contributory signals like structured data, usability, and links.
Yet relevance alone is not enough to rank a webpage at the top of SERPs. Your website needs to be a trusted source that offers reliable information and is seen as an authoritive voice. This is where the third and final pillar of SEO comes in to play: Authority. More on that in the next instalment of this series.