What is the ideal keyword density?
Let’s do a simple experiment (I like experiments), if you don’t mind: let’s take 5 unique words at random in an automotive magazine the dictionary: shelf, market, painting, section and referencing (ok they are not all chosen at random).
Then let’s look at the density of this keyword for the sites that arrive 1 ° and 11th on Google for the corresponding query. We will focus on the gross density of keywords, that is to say without giving importance to the tags in which they are found.:
plateau: first comes a Wikipedia article, 28 occurrences of the word for a gross density of 3.86%, while in 11 ° result I find a page of the DMT Web Hosting site which presents 8 repetitions of the expression and a density 0.46%
market: the first result given by Google is the home page of a classifieds site that does not have the keyword sought at all (there are 8 times the word walk, which corresponds to a density of 0.4%)
painting: once again we find in the head an article from Wikipedia which offers 62 repetitions of the searched word but a gross density of “only” 2.99%, because it is a large article. At 11 ° place, we discover the home page of the paint manufacturer V33, page on which the word “paint” is present 3 times, for a gross density of 3.16%
section: it is now a habit, Wikipedia appears in first place with and presents 53 repetitions of the term and a density of 5.07% against 9 repetitions and 1.5% of keyword density on the page concerning the Section of Historical and Philological Sciences.
referencing: it is time to finish this experience before getting bogged down inhabits, this is another Wikipedia article that ranks at the top of the SERPs whereas if I am the eleventh link presented, I arrive on the page reception of the DMT SEO agency. On the statistics side, the 1st presents 31 occurrences for a gross density of 0.82% against 51 repetitions and a density of 8.78% for the agency.
What to think of these figures? In 3 out of 5 tests, the density of expressions is lower in the highest-ranked pages, from 0 to 5%, the example of the word “market” tends to show that there is no minimum and that of “section” tends to show that a gross density greater than 5% has no negative influence.
What I notice however is that the belief, however widespread, that more than 10 occurrences of expression are assimilated to spam is completely false: the 4 pages of Wikipedia are well beyond this so-called limit.
My conclusion is simple: there is no ideal density for SEO, it’s all a question of context, semantics and HTML structure. We will see why this gross figure does not ultimately matter, without careful consideration.
Why is keyword density alone irrelevant?
Length of expressions
Overall the formula used for the density calculation is as follows: [number of occurrences of the keyword] / [total number of words on the page].
Until then everything is simple: 10 times the word “referencing” on a text of 1000 words, a density of 1%. But what about “natural referencing”? Let us assume that the expression arrives 5 times on my page, do I have to calculate 5/1000 or else calculate how many combinations of 2 words I find (ie 999) and calculate my density on the formula 5 / (1000 + 999)? But in this case, the calculation on “referencing” is also false …
The weight of words, the importance of tags
You will have noticed that I only mentioned raw density in the first part of this article because I chose to show that basic calculation is not really useful. But the tags in which the expressions are placed have a very strong impact on their importance in SEO. Thus the tool gives a gross density of 5.07% on “section” in the 4th test but it calculates a weighted density of 26.07% by giving different importance to each type of tag. Who today knows the weighting applied by search engines? In any case, this figure gives an already more relevant indication than the gross density, each SEO is free to calibrate its weighting as it pleases.
Your expression, can I use it with or without accents?
We have seen that the result of our test on “market” is surprising: while the word which does not appear at all on the page ranked first, the word “market” is represented, it even constitutes the domain name of the site. However, the meaning of the two terms is radically different and a giant like Google cannot ignore it. I would tend to attribute this result to an FQDN (exact domain name) but it proves that a complete density study must take into account this type of case.
Synonyms, plurals and conjugated verbs
It is an established fact: Google perfectly understands the different synonyms of the same word and the variations induced by plurals and conjugations. However, our density calculations did not take this notion into account at all, and rightly so: if it is easy to ask a machine to put “section” and “sections” in the same box, the terms “referencing” and “SEO” are they synonymous to the point of wanting to confuse them in all cases? Again, what the machine can provide is an indication, not a hard and fast rule.
So what should you think about working on keyword density? Of course, it is necessary to repeat the most important expressions a minimum on its pages, but this must remain natural, keyword stuffing techniques are easily detected and penalized. Yes, I think that the study of the density of the keywords can have its utility especially to detect a term too little present) but it can only be a basis of work and certainly not an end. SEO is a job that requires a lot of thinking and adaptability and if there are a lot of facilitating tools, the number of parameters to take into account is too large to let an automaton decide for you.
About the author
DMTwebhosting.com’s Editorial Team prides itself on bringing you the latest web hosting news and the best web hosting articles!
You could also link to the news and articles sections: