Possible Google Keyword Ranking Pattern – an Experiment.

[djdavejones]

Hi people,

I’m beginning to think that I may have noticed a pattern in the way that google ranks your site for certain keywords – and would like to invite you to take part in an experiment to prove this!

So for instance, 100 extra people came to my site last week through clicking on a text link in my signature on ecademy. This seemed to coincide with us suddenly going to the top of the rankings for this exact keyword.

Coincidence? Well maybe, but I’ve now seen this time and time again with different keywords. Its almost as it you get a vote for a keyword based on the number of people ‘sliding’ through to your site on the back of it.

So here’s the experiment – I’d like to invite you to click on this link "Usb werbegeschenk".

The site is based in Germany and is on page 4 for this keyword. If my theory is correct, then I’ll see a direct relation between the sites ranking for this keyword, and the number of people ‘sliding’ through to the site based on it. I’ll report any findings in the coming days.

(I’d like to point out that this site is owned by my bosses friend, not me, and therefore I'm not trying to benefit from any possible increased rankings!)

--------------------
Dave

 

[OldWelshGuy]

Dave,
Can I ask a few questions?

how does Google know what traffic you are getting to the site? are you using GA? if not how can they correlate? This has been discussed many times now, with as many people saying it does, as saying it doesn't.

I think doubt that information gathered from the Google toolbar, analytics etc is playing a part in rankings, it is just how much a part that I wonder about.

 

[awebapart.com]

It's an interesting experiment, but I think it would be a more scientific experiment if you hadn't created a new link on this thread, but sent us off (without a hyperlink) to an old site/page that already has that link on the page. That way you can rule out any improvement in ranking merely because of the new link on this thread rather than the clicks.

 

[djdavejones]

Hi, yeah absolutely Google Analytics.

I assume that Google keeps a record of how people find your site and what keywords they use if they use a search engine.

This data is probably collected anyway but only made available to you if you sign up for Analytics.

I know that for Adwords, your ad can get a quality score, if say someone clicks on your ad before that of someone above you.

 

[djdavejones]

Hi awebapart - thats fair enough, but I would imagine that links from a forum such as this are not really given any weight by Google even if it does pick up on the link.

The content of the forum is not relevent to the site, so I would guess that Google wouldn't really pay any attension to it.

 

[The Movie Booth]

As mentioned on another previous thread on this forum, there has also been some recent debate about the fluctuation of SERP's within Google, and this is certainly something which I have experienced first hand and would maybe also affect the results of your experiment. Interesting test though.

 

[F1SEO]

Yeah I think too that forum links tend to carry very little weight, although I would be happy to be proven otherwise

Your inference would mean that the more people that clicked through a link would get you better rankings. I'm interested in the results .. let's watch .. I'll click your link to be part of the experiment.

 

[awebapart.com]

The problem is you are introducing 2 factors to the experiment:

1. Extra clicks of an inbound link

2. A new inbound link

If your rankings increase what will that prove and what will people believe? At most it will prove that one or both of the factors contributed to the improvement in ranking. What factor are most people more likely to believe helped with the improvement in ranking?

1. Clicks - an area which most people do not believe, and is not common or published knowledge, hence why this experiment is an interesting one

2. A new inbound link - an area which is widely publicised as a way of improving ranking, an area which most people have heard of from all the link building SEO publicity out there (even if they don't understand what little value non-relevant forum links can be or that they fade with time)

If it was me conducting the experiment I would limit the factors as much as possible and remove 2. the new inbound link factor. That way any interesting results are more convincing.

 

[OldWelshGuy]

Personally I would approach this in a completely different way. I would have numerous test pages as well as a control. I would have people clicking on links, I would have other people clicking on search results, I would also have this being done with and without the Google toolbar installed. and all going to different sites

Site 1 linked to from forum, not clicked on (control)

site 2 as above plus people with gtb installed click link (set amount of times) posting after click to ensure integrity of the test.

site 3 as above wut without GTB installed

site 4 people with gtb clicking from SERPS

site 5 without GTB clicking from SERP's
------------------------------------------
By approaching it this way, you have a scientific (relatively) experiment and you should be able to work out the correlation between the various factors.

but that is just me

 

[deniser]

I would say it's not only the clicks that count but the time spent on the site. Having a large number of clickers who leave the site within seconds might have a detrimental effect. The number of clicks alone is too simplistic for google.

 

[OldWelshGuy]

I would say it's not only the clicks that count but the time spent on the site. Having a large number of clickers who leave the site within seconds might have a detrimental effect. The number of clicks alone is too simplistic for google.

I would agree, it needs to beorganised and set out in advance across the domains. A sudden rush, all from one link could have the opposite affect. Howevber, at least that would prove a relationship between the two, which is after all what the experiment aims to achieve.

 

[F1SEO]

well OP have you seen any ranking differences?

 

[djdavejones]

Hi all,

Thanks for your input on this one. The keywords did actually jump to page 1 of google.de within a few days of this experiment. So draw which conclusions you wish to from that!

Cheers,

Dave
----------------------
cd, usb & dvd replication