KWD experiment now underway

This test is keyword density experiment being carried out with the help of the forum. I would like to thank pro_seo for his kind offer of assistance.

Refer to this thread for the background of the experiment.

http://forums.seochat.com/seo-test-and-experimentation-81/keyword-density-experiment-165199.html

The experiment

Five unique threads will be displayed for approximately 24 hours. This will then be repeated over 5 days. Each day the previous 5 threads will be deleted.

Each day a new keyword will be added in all five unique threads (each tread will be unique each day but themed similar). The total amount of words in each thread will be 100.

The names of the threads will be - one, two, three etc day two - six, seven, eight etc..

One thread each day has kw 1 time, 3 times, 6 times, 12 times and 24 times. (same keyword different other text in each thread)

The order of the keywords are changed each day in order of threads posted.

day 1 - 1, 3, 6, 12, 24
day 2 - 24, 12, 6, 3, 1
day 3 – 3, 24, 6, 1, 12
day 4 – 6, 1, 12, 24, 3
day 5 – 12, 3, 24, 1, 6

Threads posted at same time and locked.

The keywords will be two word combinations of real terms which are not normally found together ie sports-fairy

If KWD is a significant factor in ranking the KWD of the one of these threads should rank above the others every day.

Note: Remember this is only testing if amount of keywords in page do appear to have an effect on google rankings.

The results will be published here each day near the end of the 24 hours. (Note: The results are being published after aproximatley 12 hours of being placed)

Day 1. KWD in order of current rankings a) 24 b) 12 c) 6 d) 3 e) 1
Day 2. KWD in order of current rankings a) 24 b) 12 c) 6 d) 3 e) 1
Day 3. KWD in order of current rankings a) 24 b) 12 c) 6 d) 3 e) 1
Day 4. KWD in order of current rankings a) 24 b) 12 c) 6 d) 3 e) 1
Day 5. KWD in order of current rankings a) 24 b) 12 c) 6 d) 3 e) 1

My conclusions and summation

Initial results are already out within 15 mins of opening the threads.

Click Here to see the results of the experiment.

The thread having the maximum KD [Five] is ranked at # 1. Let's wait for the end of the day to see what the final outcome is.

My gut feel says that the current ranks will stay.

Not much, but the page with a lil higher KWD has a somewhat +ve edge.. BUT hey.. thats not the bottom line, since you are not monitoring the other factors, like backlinks and other main points, I would say just in this case of Pure On-page trick, Higher KWD will have the benefit..

Day 2 test started...

Threads posted and running....

...and if you added 6, 7, 8, 9, 10, and 1000 uses... the likelihood is 1,000 will be "#1" ... demonstrating what?

...for best SEO results stuff the page with keywords??

That isn't "practical SEO".

Surely demonstrating that KWD does have an effect on this particular SERP?



True, not practical SEO, but still is an SEO factor.

Thanks for the result guys. It is surprising the thread with highest keyword density got first. We will wait for the next result. If this trend continues we can say this is still a factor but I believe there is a limit,or saturation factor in density which if exceeded does not anymore help in the optimization process.

It would be interesting to see if the effect actually becomes negative when that level is exceeded...

This is going to be fun to follow. Of course, you do realize that forum threads and web pages on product/service sites are ranked differently, right? So your experiment will only offer insights into KWD for forum threads.

Amazing really, forum threads and blog posts show up within hours on Google. I wish the same were true for web pages on sales sites!

The results of this "experiment" shows nothing tangible for SEO best practices...

If you add the phrase ONLY to:

1. Meta Description
2. Meta Keyword
3. Comment Tag
4. <noscript>

These too will show in results... that in itself doesn't mean a collective use of all of these (and only these) will generate any tangible ranks... beyond their use in experiments and using words no one will search for to find anything other than "wow KWD still works... well for everything I don't want to rank for".

If you kill the top 200 listings for all phrases... you can be sure that #201 - #999 will have examples of all Keyword Densities of text copy, only in Meta Description, only in Meta Keyword, only in Comment Tag, only in <noscript>, etc.

... but none of these will get you into the top 30.

... adding the phrase "only in the Title Element" or "only in a H Tag Page Title Element" or "only in a link to a page with 0% density" will all rank about the cirrent noted experiment examples... and used together (ONLY in the Title Element, H Tag Page Title, and a link to a page) ...will be #1 with the individual ones #2 - #4 - which means in practical application of SEO...

any service provider is wasting

Day 3 test started....

As for this test...I feel that since the kwds chosen are unique and doesn't have any competition, so the ranks are based solely on on-page stuff...and in such a case..the higher kwd density thread will always win.

This obviously doesn't state that stuff your pages with keywords, but if I am correct then this proves that for obscure keywords Google is judging the relevancy of a page by the number of times the keyword gets repeated within the document.

Again, I also feel that there should be a cut-off limit for this KD as well...if we stuff in way to many keywords and the threads are allowed to stay for quite some time, then Google might penalize those as well.

If for the same keywords, had there been pages who ranked on the basis on backlinks having the keywords in the anchors, then these threads would not have ranked the way it is doing now...coz for judging relevance..the backlink anchors would still be more important than on-page keyword density coz KD can be highly manipulated.

So long as the "obscurity" is as tested - e.g. only showing 10 results - and none of the competiting pages have a keyword in a Title or in a Link.



NEVER HAPPEN! So long as the text is visible add the phrase as many times as you wish... use the phrase like wallpaper background... where it is use 1000 times will not generate a penalty... it just won't help for ranking order.



Which is why keyword density is useless.

At some point every SEO needs to determine where the bottomline is for delivering results.

Naturally, a SEO would NEVER target any phrase on any page and never include that phrase in Titles or in Links - so this experiment isn't overly productive.

Compare your current results to a page that has 0% KWD in text copy but add the phrase to a title element with immediately go to #1... I would suggest even a phrase that is after the title cutoff (past 64 characters) say at 70+ would still rank ahead of the current results...

This is equally an "unlikely SEO ranking scenerio"... but if it ranks ahead of "just para KWD" - surely we can see just how unimportant KWD really is.

Yeah this is true, since I believe that the best SEO practices are not driven by fixed or targeted densities. It is all about relevance, and having higher densities does not mean higher relevance.

The last five posts are slightly more competative terms. (21-25)

The last five have now been posted and the order is the same. An interesting side not here though was for this thread I choose a combination of words that had about twenty other listings (as the other four had already established the main point of experiement). When you choose ‘repeat the search with the omitted results included, The thread with only the keyterm once comes number 11. Whereas the other four threads come number 1 to 4.

Anyway back to the experiment. I am quite amazed I thought that the threads would more or less display in the random order of how they were listed. On day one this view of mine was seen as the threads we in order on the page from the 24 down to the one (ie five was highest on page, then four, three, two, one) so I was surprised to see the results on day two (In the end we did not need to wait complete days as results listed and stayed stable on listed).

So what the experiment shows is that keyword density is to some extent a ranking factor for google . If it was not the threads would not have listed in order of highest to lowest keyword density every single day.

This is a surprise to me. I thought it completley irrelevant. The experiment does not show the degree of importance of keyword density when considered with all other ranking factors. BUT I do think we have proved it is one of the 200+ factors that is considered in the google algo.

Even the last day indicates that having a keyword only once on you page can be detrimental to your rankings.

As to how this data is of use to people out there trying to improve your SERPs well my advice ‘keep experimenting’. There are so many things we did not test. This was a deliberate decision . The test was only to see if KWD did have an influence on SERPs.

I would like to thank jcaputo for allowing the experiment to take place. I would like to thank pro_seo for the help in placing the threads. Without both of these this experiment would not have happened.

I would like to encourage others to use this experiment forum to help us learn more about factors that influence SERPs. So many things are just accepted as fact but when tested do not behave the way ‘the accepted facts’ say they should of.

I hope others found this experiment of some help in their understanding of how google works.

Good luck with your SEOing….


Ps – so jagnet when are we going to see you eating the keyboard