Forum Moderators: open
To be number 1 you need to have a higher PR than the others. This only works if you have your keyword first in your title and at least once in your description. I am sure it is a little more complicated than this but it is something that can be measured. How many people link to you with keyword anchor text or if you have keywords in URL just can not be measured because it has so little weight. The only time that come in to play in my opinion is if sites come to a tied with PR, Title, and description.
Example SERP
Site1: Keyword in description but is a Google directory listing and is a .org site
Site2: Has keyword in Title first word and keyword in description twice
Site3: Same as site2 but has keyword in description twice
Site4: Has keyword after 4 other words in title but has keyword twice in description.
Site5: PR7 site only has keyword in description once.
Other example:
If you type in a 2 word phrase a high PR site with just one keyword of the phrase as the first word in title will beat out sites lower PR sites that have the exact phrase in their title but after several other words.
Please sticky me with examples if you think I have just lost it. I am doing this just to figure things out. It also assumes that things don’t completely change which they may.
bottom line PR alone cannot make sure you are at top. and since google changes it algo a little every month you cannot say for sure that the thing which is imp this month will be next month also.
Well I hope this will hlep to answer few of your questions.
Deepak
Oh no, here we go again... one last time..
the lower PR site will be targeting the search query to a greater extent than the higher PR site/page.
I have top spot for many keyword terms with a PR4, if the higher PR sites wanted to they could wipe me out.
I also have PR7 site many many PR3/4 beat me for many keyword terms, so what there are gazillions or permitations, they cannot beat me for the ones I whole heartedly go far....like I said before..most times!
PR is the deciding factor if all else is equal
"This only works if you have your keyword first in your title and at least once in your description." - So your saying if I have a page rank 7 I cant get to the top without having my keyword in the title? Here is a good one and Mods please let this pass - Do a google search for - dont think so ---- that makes me chuckle like a fat man.
"I am sure it is a little more complicated than this but it is something that can be measured" -- Ok you got that one right.
"How many people link to you with keyword anchor text or if you have keywords in URL just can not be measured because it has so little weight." - Umm how long have you been doing this man? Yeah maybe if all your links are irrelevent and worthless. I shouldnt have to explain this one -
"I was not talking about the meta description. Actually I have never seen Google use it. " --- Every frame site I work on Google uses the Meta Description - Your not doing your homework.
And dude, where have you been? There has been 10,000 posts saying that there are major issues with the last update that we are all waiting to see what happens working on our sites the best way we know how.
Its kinda like you based this whole conversation on what you seen in the SERP's in your specific market (maybe a little bit further but I bet not much) and then ran in here announcing that you have seen alot of Page Rank 0 and you think that it is older data. Ummmmmmmmmm really!
"I may be way off base here" -- Yup
I have yet to have anybody prove me wrong. -- Ole'!
We have all succeded in that one -
Read read read my man and good luck ;-)
DS
If all else is equal, I'd expect any factor to be deciding (else it probably wouldn't be a factor).
I like that logic a lot, I have exact same title as you, I have exact same content (mirror if you like)
I have a PR7, you have a PR4, you are out of it,
if you don't agree why am I factoring better?
So all things not equal, you have slightly different title and content, I still have PR7 you have a PR4, you win if the search query is a better match.
So PR is not the deciding factor in this case, copy my page title etc you will lose.
If PR does not overide like minded sites, whats the point of PR?
To be number 1 you need to have a higher PR than the others. This only works if you have your keyword first in your title and at least once in your description
No. PR is not equal to keywords in anchor text of incoming links.
Here are the serps for a keyword 1 keyword 2 search, which has 1.79 million results - I'd call it a competitive search phrase.
Domenic doesn't appear to have affected these results - all the same guys are on page one, in pretty much the same order, as they have been for the past 6 months.
Result 1 PR7 Has Dmoz listing
Result 2 PR6 No Dmoz description or category. No 'keywords in context' snippet in serps - as it is Frameset parent, with no content - and this frameset parent has nothing - no meta etc. According to Google cache, 'words only apprear in links pointing to this page' - ie inbound link text.
Result 3 PR7 - No Dmoz description or category. No text on page. None. No words at all - just two title tags (keyword 1 keyword 2 is in the second title tag - which I think Google ignores), and meta tags. (meta description, meta keywords, author etc) tags. So Google is using the meta description tag - You said in #27 "Actually I have never seen Google use it" - now you have. Total ranking based again on inbound link text - google cache says.
Result 4 PR8 - Has a DMOZ description & category. keyword one is NOT on page - only in the incoming link text. Keyword 2 appears on page.
So - how does your theory relate to these flash games?
Chris_D
With everything else (including link text) equal, the amount of PR you need to overcome a better title is significant. I think I can agree that it's more than 3 notches of Toolbar PR.
Let's go back a month or two, so as to avoid the uncertainty since Dominic. We each have a 97 or 98 word page, identical except that you have the search word once, and I have it twice. If you have PR4 and I have PR2.4 then I win. If you have PR4 and I have PR2.2 then you win.
If you and I both have that word only once in the page, but I add it to the URL, then if you have PR4 and I have PR3.9 then I win. If I have PR3.8 then you win.
Therefore, I third Netizen. :-)
The numbers above are not made up, but they haven't been subject to proper scrutiny so a pinch of salt is advisable.
I have a french webpage with PR 6, wich is not listed in dmoz and 5 result on search "nom de domaine" in google.fr
The second result has PR 5 and is in dmoz.
The first result has PR 4 and is not listed in dmoz.
Well, if you understand something, tell me.
Having a dmoz listing is no magic wand - it only counts the same as any other incoming link from a page with the same PR and the same number of links on it.
Where a site ranks for a particular keyword search is down to much more than just the PR of the page - major factors include the presence of the keyword in the anchor text of links into the page, also whether the keyword is in the title tag of the page or if it's in H1 text in the page body. Look at those factors as well.
If ranking was all about PR then Google's homepage would rank number one for any keyword search.
My logic is that if all else is equal then PR will be equal too (since PR is a function of other factors). It means PR won't be a factor. ;)
not from what i understand of the theory of PR, which is at its basis, an index of the probability that a surfer would find the page by randomly following links on the Web. It is therefore orthogonal to such things as keyword density, title keywords, anchor text, and many other factors. As far as i know PR does not discriminate between say a anchor text hyperlink, a simple linked raw URL or a graphic link or any contextual information on the referring page. A link is a link, though it counts more of course if it is from a page that itself has higher PR.
I think maybe i was a bit severe when saying only when all things else are equal then PR comes into play. Of course with all the 100 and more factors, and depending on the finess of measurement, the likelihood of any two pages being exactly the same is close on impossible, unless one is a perfect copy or mirror of another. In reality the "final ranking score" ranks probably are fiarly broad, rather than very fine eg: for this query page 1 = 9.8 as compared to the second ranked page 9.7 rather than 9.87685
compared to 9.723878. This would mean that PR comes into play when pages are similar for sure, but they dont have to be exactly the same.
True that the in the original paper a link was a link. However, realizing that the probability of a surfer clicking links depends upon how they are placed or surrounded too, I won't be surprised if Google might have started considering all links not to be equal. I am newcomer to this and am just speculating but I think that links at the top and bottom of the page are more impoortant than the ones elsewhere, and links with anchor text large of in attractive colors might be more important.
