Goal
to find out how Google ranks pages and how to take advantage of that
Procedure
modelling real-life search scenario
Details
1. pick a non-existent word to do a search on. let's pick ggghhhkkk. this will ensure we do not get any other search results.
2. make 3-5 sites, each satisfying different criteria:
- title has low density of keywords ("buy ggghhhkk today!")
- there is an tag with the keyword ("ggghhhkkk properties:")
- low keyword density in a page
- high keyword density in a page
- etc, etc, etc...
3. wait till the sites get listed in Google.
4. do a search for ggghhhkkk.
Result
this should shed more light on how Google works...
You really should know more about our $1.95 web hosting deal.






Christian_SEO posted this at 18:22 — 24th December 2003.
They have: 470 posts
Joined: Mar 2001
Sounds good, but let's add two more test sites...
1) No frills site with good content, no incoming links.
2) Same as above with 10-20 incoming links, PR3 or higher.
Nielsen Technical Services
SEO Consultants
Search Engine Lists
Expired Domain Names
Join SEOBy.org - It's free
antoshka posted this at 18:32 — 24th December 2003.
They have: 30 posts
Joined: Aug 2003
we might create different types of sites, with a lot of pages, with one page, etc. the question is where should one stop? there's a lot of possibilities to consider. a lot of domains to register... lol
yes, researching PR is a next big thing... i didn't think much about it, but there are a couple of things to consider:
- what effect a linking page's PR has on a page? (approximately)
- how many tiers are considered by Google? (probably just one - that's why they need PR, to reduce calculations)
- what effect circle linking has?
in process, we might some bugs to exploit

like maybe if you put 1000 links on a page ot itself it will push your PR up to 11
You really should know more about our $1.95 web hosting deal.
Zandahar25 posted this at 18:15 — 22nd October 2004.
They have: 7 posts
Joined: Oct 2004
The bad thing is that Google tweeks it's algorithm so often that once you figure out what works, a month later it might not work anymore. I think the two things that will always be important are good content and good, relevant links.
openmind posted this at 22:59 — 24th October 2004.
He has: 945 posts
Joined: Aug 2001
Plus the fact that this thread was started in 2003 so it will now be out of date...