google SERP theory

[chillfire]

I have been watching and testing googles results for the last month or so, plus or moinus (had to get married and go on holiday in the middle....).

This pretty much meant posting to a couple of uk authoritive sites, and seeing what happened.... I added a 2 new pages to my site (chillfire.co.uk/geek). Posted some classified adverts, everyday for 2 weeks with the title of every advert including IT & Honest & support, in varying order. And added it to my V7n sig!

Serps: (on 01/Oct/07 15:21 BST)
Google.com
Honest IT support - No 2 / 15,100,000
Honest support - No 1 / 14,800,000

Yahoo.com/msn – no results in top 100

But the older pages are coming up in the searches in msn/yahoo and not in google and these pages have had NO content changes in months (neither has the site really).

msn.co.uk
asp.net development no 2 / 150,800
seo benchmark no 1 / 6,025
small business technology no 5 / 2,134,000

yahoo.co.uk
seo benchmark no 1 / 3,600,000


so...

Heres my killer theory;


Google has moved away from the search engine model that the other search engines use.
It now indexes more frequently the pages it knows about from authoritive sites. (this probably used to be DMOZ etc, I dont think it is anymore, aol ignores DMOZ, so it would make sense that google and theother s will follow suit)

It gives the new fresh content a link 'juice' score based which drops exponentially as time goes by (follows their ‘death date’ meta tag idea), so a ‘new’ page of an authoritive site will get indexed and pushed up the rankings and push down older pages SERPs, if these pages link to new pages they will be 'sucked up' the serps for a short period of time.

Big sites that have lots of ‘fresh’ inbound links and/or internal links will instantly be lifted up the serps but will have to keep adding new pages or updating major pages on their sites super frequently to stay there.

One of my major clients' competitor has been buying links left right and everywhere over the last few months, and they have strenghtned their google rankings, but have slid/stayed staic across the major engines.
They stopped linking a few weeks ago, and now their google rankings are slipping slowly back to what they were. It looks like they will ahve to keep the link freshness up to keep the SERPs.

New and old sites will only get returned in a search if there is a fresh link to that page some where in the index. Old pages with no new links will fall to the bottom.

Meaning its all about link freshness and quality of the linking pages, so crappy paid directories will give you a short term ‘bump’ in the SERPS, but over a long term campaign they will do no good what so ever as once they are indexed and given a some link 'juice' that 'juice' will deteriorate as time goes by, unless the content on the page that has the link is updates (or linked to from another authoritive site)
So if people pay a company to linkbuild, the short term gain could be great but long term could be a waste of money.

My theory does mean that there will no longer be 'index release' and that the google index is 99% fluid (with 1% or more for the authority sites it relies on), site rankings will change on a day to day (or very regular) basis, giving everyone the freshest and google 'best' results (not saying the big G is the best - I dont want to use this thread for which search engine gives best results).

It kinda makes sense, too much sense from what I have seen, but have I missed something?.

cheers for taking the time to read my theory which will probably (if I am right) be wrong by the time google wakes up reads it and changes it daily 'to do list' lol

[jelewis8]

An intriguing theory. Couple of ideas.
How do you think domain age factors into G's SERPs in conjunction with your theory?
Wouldn't contextual linking have more staying power than just a regular link? It seems like it would be difficult to gauge the maximum effect of a link if your theory is true, since any one link, once implemented, immediately goes into a "half-life" state, where it's power is constantly and exponentially decreasing.

[bond007]

It's a very interesting theory and seems to be a realistic one, too,

[jcool8]

very interesting, i usually prefer google anyway

[chillfire]

For ease of typing:
FL = fresh links
FC = fresh content
LJ = link juice
FLJ = Fresh link juice
QL = quality link
AW = authoritive website
CTL = conextual link

Originally Posted by jelewis8

An intriguing theory. Couple of ideas.
How do you think domain age factors into G's SERPs in conjunction with your theory?

Older domains still need to gather new FLJ, a perfect example would be a site that has been around for years, that has been indexed, has no FL or FC, or links to sites that have FC with FL, so it 'disappears' until a new FL is added to it and the spider finds its way to it, if the content on that page is the same as it was the LJ is low, if its a FC the LJ will be higher (depending of course on the LJ of the link in).

Originally Posted by jelewis8

Wouldn't contextual linking have more staying power than just a regular link?

Yes the FLJ would be higher on a CTL if the content/link text/linked page are all related. Links like 'click here' are will still have LJ based on the content and linked page.
However the page the CTL is on will still need to have FL/FC to keep the LJ up.

REMEMBER: in theory taking out one   or adding , or . etc will change the value of a page to a spider, so create FC.

Originally Posted by jelewis8

It seems like it would be difficult to gauge the maximum effect of a link if your theory is true, since any one link, once implemented, immediately goes into a "half-life" state, where it's power is constantly and exponentially decreasing.

Thats right but you should never aim for maximum effect of a single link, the idea is to build a level of LJ across an entire site, its like filling an ice cube tray, do you fill them all individually with one big drop of blob of water or keep pouring slowly across the whole tray adjusting the level until its all filled level.

You should never rely on a single link to build your sites value in google, it will help, but not forever, google spiders my major clients sites 200-300 time a day, so say thats 1mb of information its storing, and it claims to have billions of pages stored in its index....they only look for differences in the pages/text and keep a list of these changed pages and the pages they link to.

If it was my job to wander the internet looking for new stuff I would avoid following old paths to pages that may or may not be old. I would prefer to stay on paths less travelled to pages that nobody has been to, check them out and make note of the paths from there, do a quick check to see if I have been to those places before and go tot he ones I haven't seen and so on ...

There would be no point spidering pages that haven't changed in the last few visits, unless somebody links to them to say -this is relevant now check it out.

I dont expect to see the pages I have seen fly up google in the last 2 weeks stay there for ever.
I am not posting any new links to these pages now to guage how long it takes for them to drop away I think about 2-3 weeks will be about it.