Keyword density matters in SEO

Keyword density matters in SEO

Understanding Google’s methods of organizing search results is the first step in optimizing a page’s visibility. Of Google’s roughly 130,000 ranking factors, keyword density is a long-debated metric that resurfaced with the latest algorithm document leak.

The leak suggests that keyword density is likely fundamental to ranking a page on Google’s SERPs.

But first a word about basics and why they are important. Reliable, consistent search engine optimization uses the key factors that are likely to result in higher rankings. Focusing on the basics is both effective and efficient.

In SEO circles, keyword density is called “term frequency.” Some call it “keyword stuffing”. Either way, the goal is to increase on-page frequency for the keyword you want to rank for.

Leaked Google algorithm documents name “term frequency” as a ranking factor. Source: “Google Ranking Signals” on DixonJones.com.

‘Skin care’

Consider a highly competitive term like “skin care.” Here are the top five domains for “skin care” and the number of times that word appears in the HTML (in Chrome, go to “View” > “Developer” > “View Source”):

  1. Sephora.com – 762
  2. TheOutset.com – 165
  3. Fresh.com – 607
  4. CreamySkincare.com – 182
  5. Skinbetter.com – 596

The Outset and Creamy Skincare achieved top-five positions despite relatively fewer keyword mentions, confirming the many ways to rank a page. The other three suggest a linear relationship between keyword density and rankings.

According to Search Engine Journal in 2019, when asked if keyword density is a ranking factor, Google spokesperson John Mueller said, “…this is a pretty old metric and things have evolved quite a bit over the years…there are lots of other metrics too.”

While confirming alternative metrics, Mueller does not deny that keyword density is a fundamental ranking factor.

Google can use factor diversity to detect and disqualify attempts to manipulate its search algorithms. In other words, keyword stuffing alone does not work on e.g. an e-commerce product page. But keyword stuffing in combination with other factors does.

Top factors

Google’s top 200 ranking metrics are reported daily on Top SEO Factors, developed by Ted Kubaitis, who also makes an SEO analysis tool called Cora.

As of June 17, 2024, Kubaitis’ three most important factors are:

  1. Number of unique latent semantic keywords used [i.e., words related to the target],
  2. Number of distinct units used [i.e., an identifiable thing in a database],
  3. The domain is .com, .net. or .org.

The top two are probably fundamental, while number three is probably due to the number of TLDs using .com, .net or .org.

“Term frequency” is number 33 on the list – down 18 positions over the last 30 days – but that’s 33 out of 130,000.

The recently leaked Google documents contain many references to “concept frequency.” Read it for yourself on a page created by Dixon Jones, a developer who created a tool to research entities related to target keywords.

The Google leak is only documentation, not source code. Many notes are old with undefined words, ratios and calculations. The only certainty is therefore that Google uses, used or considered the factors at some point.

Nevertheless, based on the shared testing and knowledge of the SEO community, the leak reinforces many already existing practices.

For example, searching for “term frequency” on Dixon’s site returns six results that refer to “Tf,” which the documents define as “term frequency.” These references suggest that keyword density is fundamental to search engine optimization, especially at the beginning of a document, although the “beginning” is undefined.

Furthermore, the leak suggests that Google considers the relationship between your keyword and all the other sites on the web that use it. But it seems prohibitively expensive to calculate for each search. However, keyword density for a single page is not.

It seems difficult, if not impossible, for an algorithm to determine the current relevance of a page without at least considering keyword density. What is the alternative to a term like “computer” without using that term or a synonym a certain number of times?

Wikipedia, which ranks number one for “computer”, includes that word 1,354 times in the page’s HTML, again for all to see.

Remember that keyword density is one of many factors. Overusing it probably won’t increase rankings. However, combined with other top factors, it is likely to increase page rank, including for e-commerce products.