Keyword research is still important, but as a means to a semantic end: make sure you've covered all potential topics thoroughly and holistically if you want high rankings. Topic gap analysis assures websites cover all subjects related to an overall theme.
Two years after the launch of Google’s Hummingbird Algorithm, we now know it also has an equal or greater value for increasing your visits from Google Search. Google does not look at keywords as variables for ranking. “Topics” are a popular way to describe the broad, semantically related ideas covered on a website.
Topics are often entities (an entity has definable characteristics). Typically topics are the “mid” or “head” keywords in the old school SEO parlance. Topics are better described as thorough content, and are not a keyword metric. Google’s machine learning, for example, knows that “Rice University” is a proper name, not a university that teaches people about “rice.”
On September 22 , I’ll provide more details in a webinar with SEMrush: Find ‘em and Fill ‘em: Using SEMrush For Topic Gap Analysis.
This webinar is all about topic analysis and finding the necessary methodology to fill gaps in your topic coverage. In general, gaps are found by comparing data from your Google Webmaster Tools account (aka: Google Search Console) with a tool like SEMrush.
I’ll also show how to use your Google Webmaster data with SEMrush’s position tracker to get around Google Analytics’ dreaded not provided.
Find Your Most Authoritative Pages From Search of Your Theme
To discover topics related to Microsoft’s Xbox product, we want to isolate pages and look for other phrases Google ranks from the same page.
To find “seed” pages, we can do a keyword search on Google (better: SEMrush’s keywords tool) and confirm which pages from domains appear at the top. From this, we see that Google likes the Wikipedia entry and gives it #2 in the SERP.
As such, we can also use this page as “seed” data. Click on the Wikipedia URL in SEMrush right column and you will get the list of related keywords ranking from the same Wikipedia page to download.
Use SEMrush’s New Page Tool for Topic Analysis
Xbox.com is obviously super authoritative – one of its pages ranks #1 in the SERP.
From multiple instances of other, related phrases on the different Xbox.com pages, we find more examples of semantic relevance, otherwise Google wouldn’t send traffic there. If “Wizard of Oz” appears on two pages from the domain, at best Google will rank the term just once.
Remember, Google’s goal is to answer searchers’ questions. It ranks the other terms on a page if they are related based on its machine learning. If “Wizard of Oz” is not appearing on other authority web pages, on other websites, it will not send people to a page that’s not about Oz.
The “authority” also comes from a combination of links to the page and other ranking signals. A very popular website about the Xbox will be ranked because it has the linking “vote of confidence.”
What are the chances that “Wizard of Oz” will also appear on both Xbox.com and Wikipedia pages about the product? Very, very low. If it appears, there’s relevance. You may be an Xbox expert, but what if all these websites are discussing the greatest game ever planned for Xbox: “The Wizard of Oz” fantasy, gaming extravaganza? If you missed the news, the topic analysis finds it for you.
The SEMrush Pages tool lets you sort by number of keywords ranked for a page. We find the top pages have 754- 1250 other words ranked from the page that are part of SEMrush’s database of 40 million keyword phrases. These are pages that are the most authoritative, and by looking at the URLs we can see exactly what they cover.
To find co-occurrences, we download the Excel sheet for each of the pages and put all the phrases into a single worksheet.
The occurrence formula is in the below screenshot.
Now we can use common SEO metrics like search volume, but we also see how often the word is used on the highest authority websites!
The highest number I got for this one was phrases and topics appearing on each of three pages and none on all four.
There were about 100 of the related phrases and topics with a score of three. Total related phrases was 4,195. However, the strength of association is too low to look carefully at all but the ones with the three occurrences.
Ready to learn more about finding your content gaps? Join me on my upcoming SEMrush webinar on September 22.