The Subtleties of Keyword Density and Stacking

[Komodo Tale]

I am hoping that some people with experience can address Keyword Density and Keyword Stacking (using the same keyword over and over again on a page to increase density).

In content rich sites there is a tendency to use the same Key Words and Phrases repetitively on a page so as not to dilute them. Other times, important words can have a lack of synonyms or alternate terms, which forces stacking.

On the other side of the balance beam are sparse websites that employ very little text around their Key Words and Phrases.
For this discussion, let’s assume that both types of sites have the same Keyword Density. One site is content rich and user friendly, but the language might read a little unnatural. The other site reads more naturally, but there is little content to read.

The obvious examples that I am thinking of are shopping sites, though I’ve seen other examples of this as well, and it seems that very often less content is rewarded more strongly than rich content by Google and Yahoo! (MSN seems more balanced). I would think that the opposite would be true so I am wondering why this is and what guidance you can provide?

Also, perhaps a tutorial about how search engines use Keyword Density?

[Phoenix]

I only problem I see with the question is that it doesn't factor in many other elements that could be of importance in ranking those two documents. Primarily links and the layout of the html (i.e. where the content resides in the document) and on page elements such as headers, bold tags and so on.

Per your example, the site that has richer keyword density, apt is more difficult to read could rank just as easily as the site with little content. I think its also helpful to abandon looking at keyword density as well, as its use is to draw conclusions on the success of how fit the document is to rank highly for a keyphrase. These days I don't know many SEO's that use keyword density in their optimization. Its a waste of time basically. Shopping sites can have great sucess but for them its a more fine tuned task in order to rank for more competitive phrases. Such as more agressive linking and using keywords in the right order and in the right places. Really good internal linking is also prime for large retailers in helping those product pages listed.

A helpful cross examination of spam/Adsense sites and those sites which use autogenerated content might be helpful. Programs are used to generate the content and as such the language isn't as good. However, as you can see in any search engine these sites continue to rank highly for many terms. Sites that additionally scrap search results and use it as content is another example.

With keyword density aside, maybe you could restate the question, in regards to the differences between ecommerce sites with little content and those sites with a lot of rich content. The debate about these two types of sites has gone on for a long time. The concensus is that very good linking can pretty much compensate for any negatives those sites might have.

[Komodo Tale]

Your assessment is fair enough, however I am more interested in a "when all else is equal" scenerio. Rather than looking to other ranking variables to overcompensate I'd like to flush out keyword density and stacking.

Perhaps I could devise a test. I have a parked domain on which I could place two test pages, one short and one long, with the same nonsense keyword and same keyword density. Then I could ask forum members to link identically to both. We could then see which page ranks higher in different search engines.

Would such a test interest people? Would you -- yes you, the person reading this right now -- would you participate?

[g7submit]

While I wouldn't mind participating, I think if someone has has tried, tested, and has tweaked results in trying to find a solution in the past in a particularly controlled environment would feel pulling back the hands of time is not best route.

I have worked on various ecommerce (with dynamic URLs) and static websites in the past and now know better what works and what doesn't. As suggested by Phoenix, external links do compensate for any shortfall in a site's structure. But as you wanted to verify, what if the site structure is the same except the content, well, from personal experience, keyword placement within a content is the major influencing factor. I have seen this repeatedly.

If I may repeat what Phoenix has mentioned again, however, that IBL (keyword-specific anchor text) still compensate and influence the rankings more than on-page optimization. Also experiments shows that ecommerce sites with mod rewrites are as good as any static site. What you are trying to discover is most glaring in flash and static sites.