Monthly Archives: April 2016

Google Ranking Factors Overview

We’re talking about Google ranking factors because Google is by far the dominant search engine; however, these same factors apply to other search engines as well. All of the major search engines today use the same tools & factors to rank websites. The difference is how much weight they put on each factor, how much data they have on those factors, and their ability to remove spam from the results.

When we talk about ranking factors, we’re talking about things that directly influence how the search engine ranks a page of your site. Thus we are not talking about tactics, which are the methods you use to influence a factor.

Thus links from social media is not listed because it’s not a meaningful ranking factor; however, leveraging social media to make a ranking factor happen (like getting a link from a news site whose editor saw your tweet) is a viable tactic.

The goal of this section is to give you a good overall sense of what the ranking factors are. Later we’ll talk about how to optimize each of these factors, and how much attention to spend on each.

Broadly speaking, there are three categories of ranking factors:

  • Topicality (is the page about the topic the user searched for?)
  • Authority (which topically relevant page does the democracy of the web think is best?)
  • User Metrics (which page do users seem to like the best, based on their behavior?)

These three factors combine in uneven proportions to determine relevance. Google’s goal is to deliver the most relevant search result.

Authority Ranking Factors

Topicality factors alone aren’t enough to rank, and the introduction of authority factors is what separated Google from other search engines at the time, and allowed Google to gain market dominance. Let’s take our “volcano sledding” example:

"Volcano Sledding" has 170.000 Google results

170,000 results for “volcano sledding”! All of them are topically relevant to some degree, so which one should rank first?

For most search queries there are hundreds of thousands, if not millions, of pages on the web that are topically relevant. There are thousands that are highly topically relevant. Authority signals are how Google decides which of the topically relevant pages to rank first.

Authority signals are almost entirely based around links. The idea is that the topically relevant page that the most sites link to is probably the best page on the web for that topic. Of course it’s not just about the number of sites, but how authoritative those sites are as well: Google cares a lot more about a link from CNN than it does a link from your college roommate’s blog.

As search engines have grown more sophisticated, they have become more nuanced in how they understand different aspects of links, and today there is a whole host of different aspects of links that count as authority signals. Major authority signals are

  • PageRank of linking pages: or the raw link juice, Understanding PageRank is important for SEOs. You can think of it as a measure of the raw numerical authority that flows from every single link into your site.
  • Number of linking domains: the more sites that link to your page, the better. The raw number of linking domains is the single largest ranking factor in SEO. One link from 10 sites carries far more authority than 10 links from one site. Some people dive deeper, and look at the diversity of the IP address C-blocks of linking sites. You do not need to worry about this — this is only a concern for people doing blackhat tactics who artificially create huge numbers of links from fake site networks.
  • Position of link: a link in the footer of a site counts less than a link in the header, and a link within the actual unique content area of a page is better than either. Similarly a link toward the top of the content is better than a link down at the bottom.
  • Relevance of linking site: a link from a page about a similar topic is better than a link from an unrelated page. So if you’re trying to rank a page for the term “artisan snowflakes,” a link from a page also about snowflakes is more valuable than a link from a page about sports scores, or a link from your little sister’s blog. In fact, links from completely unrelated pages may not be counted at all (they usually are though, but only a wee little bit).
  • Rel=nofollow: this is HTML code that can be placed on a link that causes the link to pass essentially no SEO value. This is typically used on all blog comments, social media sites and many forums.
  • Link frequency: in more competitive markets, the frequency or velocity of links also matters. If one site is getting a new link every other day, that is a strong signal when compared to a site with more links, but that hasn’t gotten a link in the past two years. For lower competition terms, frequency doesn’t matter much.

Topicality Ranking Factors

Topicality ranking factors are elements that improve Google’s understanding of what your page is about, and whether it’s topically relevant to a given search query. For example:

Google search query for "volcano sledding"

If you’re searching for “volcano sledding”* Google will look at the index and try to find pages that are about volcano sledding. Topicality factors are what help Google determine if the page is about volcano sledding, and how deeply about it: does the page just reference volcano sledding in passing, or is the entire page dedicated to the topic?

Topicality factors exist almost entirely on-page, and it’s very easy to maximize topicality (what I usually refer to as SEO kindergarten).

  • URL: do the keywords appear in the URL
  • Title Tag: do the keywords appear in the title tag
  • Keywords in the content, as well as synonyms and related terms (co-occurance)
  • Word count of unique content — higher word count of quality and unique content is better
  • Anchor text of links pointing to the page; anchor text from external sites carries much more weight than anchor text from other pages within the same site
  • Relevance of linked sites: the topic of web pages linking to the page can provide minor topicality signals — as can the sites that you link out to
  • Minor factors: there is some debate about very minor factors, like H1 or H2 tags, text, etc. There is virtually no correlation between these factors and rankings, and you can safely ignore them.

It’s worth noting that there appears to be a kind of cap on topicality. I once did a series of tests across a dozen websites, where I removed relevant keywords entirely from title tags or built pages without keywords in the title tags.

Most pages did not lose a single rank — even after months — from removing what is generally considered one of the most important ranking factors. But for some pages their rank plummeted.

The correlation I discovered was that for pages that were well optimized, the title tag didn’t matter at all, and for pages that were not well optimized for a given keyword, the title tag was hugely influential. I then extended the test to other on-page factors and found the same thing in every one.

It appears that there is a cap to how topically relevant a page can be. And this makes perfect sense when you think about it: at some point Google looks at a page and says, “Hey, this page is completely dedicated to volcanoes; that’s all it’s talking about, volcanoes all day.” Once you hit that point when you’re page is considered 100% about the topic, adding more topicality signals isn’t going to make the page more about that topic; it can’t be more than 100% after all.

It’s also worth noting that the only factor that does not appear to have a topicality cap is the unique word count. Even when you reach the point that title tags and URLs no longer improve your topicality, adding another few good paragraphs can still help — though there are diminishing returns.

* I just made this phrase up as a funny example. But when I entered it into Google I discovered that volcano sledding is a real thing. Here’s a photo from this page on the topic.

Volcano sledding. Yes, really.