Keyword density: why the number alone brings little
What is meant by keyword density, why rigid target values help little and how you create relevance better.
The topic often decides whether content is only available or actually meets the search intention. Usually, the biggest lever lies in better tailoring, clearer utility and less in mere amount of text.
What is meant by keyword density, why rigid target values help little and how you create relevance better. This wiki page arranges the topic practically and shows, When priority arises, which mistakes are often and how you derive concrete measures from them.
A fixed keyword density is not a valid ranking factor. Crucial are search intentions, helpful content, clean information architecture and naturally set terms instead of percentages.
Example: A high quota arises from repeated navigation
A tool often reports the word “shoes” because it occurs in every level of navigation. Compare the main content with the entire extracted document. Different text ranges and word counting methods can provide significantly different percentages.
There is no ideal target value derived from this quota for each page. Use the evaluation to detect eye-catching repetitions, not to mechanically fill a text. Complete content and comprehensible formulations remain the actual editorial task.
What is keyword density important for?
It makes sense to everyone that it has to be good if a key term appears more often and the crawlers of Google in a text find it good. In the early days of search engine optimization, around the year 2000, many optimizers therefore overloaded the texts with the search terms that seemed important to them. They were partly quite clever: a common method called for the elementary word, which the user is looking for, to be placed right at the beginning of a text, then at the end, in all headings (best again alternately at the beginning and at the end) as well as in the text well distributed.
A common SEO hypothesis was that the crawlers always browse a text from the front and from behind, section by section and headline by headline. No one but the programmers of search engine algorithms knows exactly whether this is true. In any case, the keywords were too common in most texts, which leads to the concept of keyword spam as an unauthorized method of Black hat SEO led. Google is said to have sometimes punished such sites if they had a keyword density of 10% and sometimes even more.
However, it should be noted that the factual evidence for such penalties is lacking. Nevertheless, from 2010 at the latest, the principle was established that the keyword density should be 3 – 4%: no more and no less. Optimizers adhere to this until today (2021) and give their authors appropriate instructions.
Why do search engines need a certain keyword density?
You classify the SEO guide value of keyword density in the evolution of search engine technology. Of course, they need Googlebots or other crawlers (other search engines or social networks) provide an indication of what the focus of a text is. This clue is provided by the decisive keyword. However, how the algorithm of keyword recognition works exactly must remain a mystery. After all, many other words also appear very frequently in a text without being considered keywords.
It is noteworthy that search engine operators keep very low on the topic of keyword density, when they willingly provide information about various other SEO topics. At the same time, key terms are used in online texts with almost fanatical meticulousness. Text agencies offer their customers keyword counters for this. The authors now have to put the corresponding words exactly in the given number in the text (often from - to), which can be very exhausting and seem highly superfluous.
There is a suspicion in author circles that keyword density is a carefully cultivated myth. On the other hand, it is obvious – see above – that the crawlers need some terms to look for, and that these terms should appear in a meaningful cluster in a text – not too often and not too rarely.
What else needs to be considered in terms of keyword density?
If you take the keyword density very seriously, you must note that it also includes keywords in areas that are invisible to the reader. These are words in the HTML code of a link, in source codes as well as alt and title tags.
WDF*IDF as a replacement for keyword density?
The keyword density has been doubted for some time regarding its importance for SEO as described. Around 2012, the SEO scene rediscovered the weighting of terms using WDF*IDF. This is the document-specific word weighting according to the linguist Donna Harman, for which there is a rather complex formula. It was actually developed for scientific documents and is intended to give the significant words and terms (word combinations, terms) of a document the value they deserve for the document in question.
This means, in short, that in a text that deals with apples, their varieties and their processing, but in which they are also occasionally compared to pears, the terms “apple”, “apple variety” and “processing of apples” should appear much more frequently than the term “pear” and everything related to it. How much more often it has to be is inferred from the context of the text and can be calculated with the WDF*IDF formula, which is not to be mentioned here: it appears to practically active authors as nonsense in pure culture and completely superfluous headbirth, according to which no one can write a reasonable text.
Donna Harman probably had nothing better to do when she invented it. I prefer keyword density.