PR - Just a hype?

[PixelStreamed]

Okay, so we know everybody is interested in receiving a high Page Rank for their website pages. But does anyone actually know the benefits of having a high page rank? It would probably help (a little) in Google's algorithm in the SERP's and maybe make your website seem more valuable and trustworthy to visitors, but other than these two things what else is a high PR good for? I haven't thought of any other benefits a high PR would be good for; it seems like it's more hyped up than it should be. So what's the PR craze all about?

[seobook]

it makes other people want to trade links with you more. many of these people are not worth trading links with though...

[newsphinx]

make your website seem more valuable and trustworthy to visitors

Average searchers, other than webmasters, may not have idea what is PR. :D

[Nick W]

As seobook hinted at, its a prestige thing. It makes you a target for link requests but also for non-recip incomings.

I dont really follow PR but surely it's still the case that a PR9 would out-rank a similarly optimized site with PR6 right? - But around the 4-6 level I dont think it counts for very much...

Nick

[webcertain]

I think PageRank is fascinating for a huge amount of webmasters simply because it is measurable, and what's the point of calculating a boring cpa or even roi when you can see the results of your efforts just by opening your browser :D ..

[webcertain]

I dont really follow PR but surely it's still the case that a PR9 would out-rank a similarly optimized site with PR6 right? - But around the 4-6 level I dont think it counts for very much...
Hi Nick, PageRank is part of the puzzle, and no way you can be sure that a high PageRank is a guarantee to be in the first serp.. An example in the news:


football tickets euro 2004 in Prog (http://www.webmasterbrain.com/prog/search?hl=en&lr=&ie=UTF-8&oe=UTF-8&q=football+tickets+euro+2004&num=10&btnG=+++Search+++)

[Nick W]

Hi webcertain, yeah, i know what you mean, i just meant in the most general terms.

But, point well taken ;-)

Nick

[rustybrick]

PageRank is part of Google's overall PageRank algorithm. How much weight PageRank currently has, only Google knows. But I would guess that if all other factors are equal between two sites, the site with the higher PageRank would rank higher.

seobook is correct in saying that it is most probably easier to trade links with friendly sites if your PageRank is higher then the average. But many link traders are not looking at PageRank these days. Many are looking at the number of pages they can get their links on. They are also looking at if the pages distribute link value to other pages.

[webcertain]

They are also looking at if the pages distribute link value to other pages.
Hi Rustybrick,
distribute link value, that would be something like pagerank / number of outbound links no?

[rustybrick]

Hi Rustybrick,
distribute link value, that would be something like pagerank / number of outbound links no?

I am not using an equation to define what "link value" is. IMO, link value is inclusive of PageRank, anchor text, # of links, the page content of the page linking to you and more.

[webcertain]

Ok, thought you were talking about link value on a PageRank point of view, ie mathematical. but true enough, anchor text and context of the page are important, but not measurable (yet?)
Cheers,
Johann

[David Wallace]

... what else is a high PR good for? I haven't thought of any other benefits a high PR would be good for; it seems like it's more hyped up than it should be. So what's the PR craze all about?
If you are selling advertising or placement on your site such as with a directory or some kind of information site, the higher the PR you have on your pages where ads or placement will take place, the more attractive it is to advertisers looking to increase their own PR.

Other than that, the average person who isn't marketing site probably doesn't have any idea what PR is all about.

[Incubator]

Does PR have any tie into the toolbar at all?

What I mean does google give more factors to your site, if it is search through the toolbar and not straight at www.google.com.
I can see who google would benefits from this for ...lets say...thier tracking information .

For example if 100 people searched through thier google toolbar for "bluewidget" and everyone landed at google and clicked on the first site.....
would that carry any weight on the user end of the toolbar, that was being used across a percentage of the 100 surfers ?

:confused:

Cheers
Wayne

[Anthony Parsons]

I think the only thing PageRank is good for is a marketing tool. Those who know no better quarel about those low PR values and links, who cares, but the real importance is the time saving factor finding those sites that do hold the authority.

Unless your talking about PR8 +, then PR value has no real importance, not even for rankings, which is proven time and time again.....PR4 ranking over PR6 etc. You certainly can't use it for traffic analysis, as you can have a PR7 with 5000 monthly visits and a PR7 with 500,000 monthly visits.

What gets me is people's attitude towards PageRank. It is a measure of Googles importance only. That doesn't mean squat to Yahoo and MSN who take up the other two large chunks in providing results. If MSN take over the search lead in a year or two, then what? Will we have MSNRank or something like that, that the world will fall over and start comparing one another against? Yuk....

I think the sooner we can get into peoples minds that it means very little, the better we will all be.

[jbgilbert]

..... It is a measure of Googles importance only. That doesn't mean squat to Yahoo and MSN.....

..... I think the sooner we can get into peoples minds that it means very little, the better we will all be...,..

I'm for that!

The FIRST thing people need to understand about PR is that THEY DON'T REALLY KNOW WHAT THE PR VALUE IS!!!!!! The only visibility into Google's REAL PR is the silly 0 to 10 we are allowed to see.

The PR that we see is only a crude approximation..... PR is achieved through linking..... PR is the "cart", quality links are the "horse".

Stop putting the cart before the horse.....

[nuclei]

As seobook hinted at, its a prestige thing. It makes you a target for link requests but also for non-recip incomings.

I dont really follow PR but surely it's still the case that a PR9 would out-rank a similarly optimized site with PR6 right? - But around the 4-6 level I dont think it counts for very much...

Nick

Incorrect. A PR4 page optimized identically to a PR9 page can easily outrank the PR9. It is a matter of the PR4 simply using the keyword he wants to beat the PR9 for in his anchor text where the PR9 does not.

PageRank does not have much if anything by itself to do with rankings.

[webcertain]

PageRank does not have much if anything by itself to do with rankings.
I agree, PageRank has no direct consequence on the rankings, but some of the links that composed the PageRank may well be one of the main factor...

In fact, PageRank is a very poor indicator of the "ranking ability" of a page..
But it's the easiest indicator to monitor..

[simon@bizcomservices.co.u]

I think that PR is overhyped, and interestingly posted a similar thread on the high rankings forum called "Is Pr Overated Or What? (http://www.highrankings.com/forum/index.php?showtopic=6944&hl=)".

Some people chase PR thinking that it's the ultimate goal, but most users don;t know about it, or care about it, and I've not seen much evidence to suggest that it has very high weighting in Google's algo.

I've seen PR3 ranked higher than PR6 for certain phrases (just search for "uk website promotion" in Google and look at #1 and #2 PR.

[Anthony Parsons]

I would like to add one little thing to this. PR is a marketing tool, however; on the other end, PR does play some role in rankings. Think it doesn't?

PageRank is the algorithm that "Google" uses to measure the quality of a sites backlinks. PageRank shows an authoritive rating (in simplest terms) for a given site for Google only. If your niche relates to that particular niche with a higher PageRank, then it does share some effort towards demonstrating the authority of the site your link is being placed upon that will pass some authority too your site (this all being related niche content).

So, PageRank does play some part in a sites ranking, because when you start talking same niche sites linking, PageRank (authority) is passed to help that site being linked in the rankings for "that" niche. So it does play a role, just not how 95% of the web think. Most of us here know better. Yes, it does mean little in the scheme of things, but does play a role as the algorithm measuring linking. Makes it a damn site easier to find what G finds as authoritive and what not in a relevant niche. That's handy and time saving.

When you get over PR8, relevant just goes out the door and PageRank then does play a factored role in rankings. When you're getting that high (authoritive) you will rarely see a PR8 site that is directly targeting the same term go over a PR9. It is only in the lower roles where people really get confused about linking to only PR4+ and so forth. If the site is relevant and contains great content, link to it and link from it.

Also, on the note of PR8+ sites, that type of authoritiveness generally is a good indicator that reflects within Yahoo and MSN rankings also. If you can get any PR8 site linking to you, and only you, your site will be improve dramatically within rankings. That is when PageRank has effect as a PR value.

[webcertain]

PageRank is the algorithm that "Google" uses to measure the quality of a sites backlinks.
Measure the quality? That's the issue here, what do you mean by quality?
PageRank: PR(A) = (1-d) + d (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn))
I would rather say quantity of vote, for quality, IMO, it's more a question of anchor text, content of the page where the link is..
Cheers,
Johann

[nuclei]

Also, on the note of PR8+ sites, that type of authoritiveness generally is a good indicator that reflects within Yahoo and MSN rankings also. If you can get any PR8 site linking to you, and only you, your site will be improve dramatically within rankings. That is when PageRank has effect as a PR value.

This has not been any evidence of this at all. In fact Many pr4 sites out rank pr8-9 sites. Also telling people that a single link from a pr8-9 will "dramatically" improve rankings is a common misconception. One link is still one link. PR9 or not. It does not do all that much to the SERPS. Now if you are talking a sitewide link on a PR9 site where the inside pages are all pr7-8 thats a different story.

[nuclei]

Measure the quality? That's the issue here, what do you mean by quality?
PageRank: PR(A) = (1-d) + d (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn))
I would rather say quantity of vote, for quality, IMO, it's more a question of anchor text, content of the page where the link is..
Cheers,
Johann

Exactly. Anthony said a few things in that post that were a little off. One, PageRank does not really have anything to do with a "site". It is URL specific.
Also pagerank does not care about quality and it is not a measurement of quality of links. It is googles estimation of a "pages" importance, not a links.

[Anthony Parsons]

This has not been any evidence of this at all. In fact Many pr4 sites out rank pr8-9 sites. Also telling people that a single link from a pr8-9 will "dramatically" improve rankings is a common misconception. One link is still one link. PR9 or not. It does not do all that much to the SERPS. Now if you are talking a sitewide link on a PR9 site where the inside pages are all pr7-8 thats a different story.

As I said and others, lower PR means nothing, go over PR8+ does. I know this as a fact, as I scooped a deal for a client on a PR9 .org site, only link out, and smack damn, that site come from nowhere to page one for 16 million and above allintitle search terms. It does work mate and anyone who has competitors in the highly competitive arena will tell you that Nuclei.

Now, for the other engines, PR8+ does not give you any sort of guarantee for anything, nor Google for that fact if you don't know what you're doing, but as I clearly stated, it is a good method to show you what is actually authoritive and what isn't.

[Anthony Parsons]

Exactly. Anthony said a few things in that post that were a little off. One, PageRank does not really have anything to do with a "site". It is URL specific.
Also pagerank does not care about quality and it is not a measurement of quality of links. It is googles estimation of a "pages" importance, not a links.

Nuclei, a quality link in my perception is the quality of the page the link reside upon and the page the link points too. That is what G measures.

[nuclei]

I know this as a fact, as I scooped a deal for a client on a PR9 .org site, only link out, and smack damn, that site come from nowhere to page one for 16 million and above allintitle search terms.[/b]

Please cite an example. Nobody else has ever been able to yet. Never has a single link from a pr9 site beaten out a similarly optimized page with a bunch of lessor pr links with proper anchors in my experience.

[dannysullivan]

This has not been any evidence of this at all. In fact Many pr4 sites out rank pr8-9 sites.

Agreed -- or to be perfectly clear, there are many URLs with a lower PR score that will outrank those with higher ones. I'm always writing about this, and it's one reason I love the new tool prog (http://www.webmasterbrain.com/prog/) that webcertain mentioned earlier. Do a search there say for cars (http://www.webmasterbrain.com/prog/search?hl=en&ie=UTF-8&oe=UTF-8&q=cars&num=10). You'll see Jaguar at PR6 will do better than CarsDirect.com with a PR8. Other examples are easily found on that page or with other searches.

But I would guess that if all other factors are equal between two sites, the site with the higher PageRank would rank higher.

Again, while it's always hard and easy to slip, it's a page-by-page, URL-by-URL thing. So all things equal, two pages exactly the same in ALL ways, would the higher PR one do better. I'd very, very hesitantly say yes. But this is also practically impossible. The two pages, to be EXACTLY the same, have to have all the same exact links, same body copy, etc. Given this, they'd end up with exactly the same PR score.

As I said and others, lower PR means nothing, go over PR8+ does. I know this as a fact, as I scooped a deal for a client on a PR9 .org site, only link out, and smack damn, that site come from nowhere to page one for 16 million and above allintitle search terms.

Anthony, I'm going to guess that this link had one or more of the key terms that page was ranking for in it. Was that the case. If so, yeah -- a single link like that could have that impact. But, a single or just a few links from relatively low PR-ranked pages might also have the same impact, depending on the term. That's happened in several cases of Google/link bombing.

Okay, so we know everybody is interested in receiving a high Page Rank for their website pages. But does anyone actually know the benefits of having a high page rank?

As others have said, it might help you with link swapping. The most dramatic thing I'd say is does for you is let you pass on high PR to your own internal pages or those outside your site you care about. If you are deemed important, then a good CONTEXTUAL link out to another site I could have a good impact for the terms in that link.

So what's the PR craze all about?

In my view, it's because there's a giant disconnect between the idea of the context of links being important from the value Google's little meter shows. People like numbers. It's easier to say, "This must be a good page, it's a PR9" than to say, "Do a search for 'cars' on Google. The pages in the top results are likely more helpful to you because Google itself is telling you they are both important for that term and important as deemed by its ranking algorithm in general."

This past article I did back in 2002, Google Sued Over PageRank Decrease (http://searchenginewatch.com/sereport/article.php/2165111) talks a lot more about the PR madness from when it was really starting to grow, plus has my golden rules that I personally think are still helpful. This one, Google PageRank, Meet Yahoo Web Rank! (http://searchenginewatch.com/sereport/article.php/3334891), is sort of an update/fear that Yahoo's own meter will just make matters worse.

[Rob]

Personally, I use PR as an indicator when building links but that's about all I use it for. I don't dismiss sites if they have a lower PR, especially if they are relevant links to my site.

That's the key here. While a High PR site may help improve your rankings by increasing your site's PR, if the link is from an unrelated site, it has less effect in my experience.

i.e. a link to Microsoft from Joe's plumbing is worth less than a link from another software vendor, even if Joe's site has a similar or even slightly higher PR value.

I prefer to look at what the site is about and see if the link to (and from) my site would be a benefit to my site's customers regardless of the PR.

When it comes to link building this is what it should be about. Helping the engines build in the relevancy and not exploiting it.

Just my 2 cents

[polarmate]

For us, PR is of low concern. Of primary concern are conversions and RoI which are not related to PR. We were doing well in the organic listings even with a PR4. We've been at PR6 for a while now. The impact of PR on RoI was negligible when we moved from PR4 to PR5 to PR6. PPC strategies mattered more with rankings coming in a close second.

[PixelStreamed]

As I said and others, lower PR means nothing, go over PR8+ does. I know this as a fact, as I scooped a deal for a client on a PR9 .org site, only link out, and smack damn, that site come from nowhere to page one for 16 million and above allintitle search terms. It does work mate and anyone who has competitors in the highly competitive arena will tell you that Nuclei.

Now, for the other engines, PR8+ does not give you any sort of guarantee for anything, nor Google for that fact if you don't know what you're doing, but as I clearly stated, it is a good method to show you what is actually authoritive and what isn't.

i can backup Anthony's statements....we tested this with a brand new registered domain. as soon as we brought up the domain and had a pr 8 site link to it the PR of the new domain jumped to PR 7 within two weeks, but what exactly the PR 7 is good for (other than other sites wanting to trade links with you) I have yet to find a solid answer. To me it seems the only reason PR is getting a name for itself is because of all the hype from people wanting to get a high PR. other than that it is basically worthless. it's been shown that PR doesn't play that big a role in rankings because there are tons of low PR sites that outrank really high ones. that is why i suggested in this post that PR might be just a bunch of hype and if people ever get off the PR hype it will basically become worthless.

Yahoo already has plans of making its own type of PR ranking system...so i guess the PR hype is here to stay for awhile.

[nuclei]

I love the new tool prog (http://www.webmasterbrain.com/prog/) that webcertain mentioned earlier.

Funny you should mention this. Not 2 weeks ago he had the problem where someone had stolen his code and was using it rather widespread. He was ready to throw in the towel when I asked if he wanted my help with the matter. He agreed. The other site was taken down the next day and he is back to doing well, as he should be.

I hate when fellow coders have to go thru that :)

[PixelStreamed]

quote:
This past article I did back in 2002, Google Sued Over PageRank Decrease talks a lot more about the PR madness from when it was really starting to grow, plus has my golden rules that I personally think are still helpful. This one, Google PageRank, Meet Yahoo Web Rank!, is sort of an update/fear that Yahoo's own meter will just make matters worse.



just saw your post about the Yahoo Web Rank....good article...

[AussieWebmaster]

Beware of the scams.... 301 to a PR8-9 site and once spiders recognise the other page and the ranking hits the high number... change the page to an ad buyer and off you go>

[Anthony Parsons]

Measure the quality? That's the issue here, what do you mean by quality?
PageRank: PR(A) = (1-d) + d (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn))
I would rather say quantity of vote, for quality, IMO, it's more a question of anchor text, content of the page where the link is..
Cheers,
Johann

That is exactly my interpretation of measuring the quality also webcertain. Anchor text placed upon a similar or contrasting content rich page = quality. Sorry about that confusion, that is what I deem as quality of linking. I think most understand about "quality" of linking.

[Anthony Parsons]

Anthony, I'm going to guess that this link had one or more of the key terms that page was ranking for in it. Was that the case. If so, yeah -- a single link like that could have that impact. But, a single or just a few links from relatively low PR-ranked pages might also have the same impact, depending on the term. That's happened in several cases of Google/link bombing.

Nope, not this time round. The site is a non-profit charity organisation that took a kind "undisclosed" monthly sum from my client, donated for exclusive rights with no further site linking. I had heard of this done before and done it myself with PR8, but a PR9 is another adventure again. The clients site is really just a general all purpose service.

Dan actually had a gander at a site similar in nature not long ago that we are just about to do the same thing for. The site has all the PR1 - PR6 backlinks as with their competitors, but the site in question that I am advising now has an allintitle of 19 million for one of its terms. All the top 10 sites currently listed for that term have atleast one PR8 page backlink, not even relevant to the field, pointing at them.

I totally agree that PR means nothing, and really means nothing all through the range to me, however; how I look at it from PR8 above is the authority that site carries. If Google gives it a PR8, then the site has some authoritive backing from G, hence which can pass lots along exclusively was my theory. The theory works. I am actually going to openly test this theory a little later on down the track if I can find someone with a PR8, PR9 and PR10 if possible, that is willing to put a link on their site pointing to some new domain with nothing to do with it and record the results for some highly competitive term. Clients sites I cannot do that, but if we could get some volunteers who don't mind giving their page for testing purposes, then we could disclose those results openly with all facts attached and verified. I am working on that as a project for this year.

Hint Hint Danny......Unless you or someone else know someone with a site who would allow it for testing purposes at that level now for no cost (testing purposes only on specific domain created for the purpose). I want to test sites with no outbound links then "x" outbound links and so forth.

[nuclei]

If Google gives it a PR8, then the site has some authoritive backing from G, hence which can pass lots along exclusively was my theory.

Anthony, I may be missing something or not reading what you are trying to say, but once again, pagerank is 100% dependant on inbound links. It does NOT care where they come from or if they are even from relevant websites or not. Googles giving a page a pr8,9,10 has nothing to do with anything at all other than the number of IBL's to it. It is not because google feels it is an authority site or that google even gives a rats rump feathers for the page. It is only about links. The only relation it has on rankings is a little added weight given to anchor text in links.

This is just how pagerank works. Nothing more, nothing less.

Oh and this is not just "from reading an article". PR8 sites are nothing new to me, I have had several and have done my own tests.

[seobook]

I would be willing to host whatever site was needed for a test that was going to have PR8 + links pointing at it. I have reserved many domains just for this purpose.

I still have not done anything with www.whitehatseo.com (other than put the reserve for high PageRank inbound link test statement on it) :D

[Anthony Parsons]

Anthony, I may be missing something or not reading what you are trying to say, but once again, pagerank is 100% dependant on inbound links. It does NOT care where they come from or if they are even from relevant websites or not. Googles giving a page a pr8,9,10 has nothing to do with anything at all other than the number of IBL's to it. It is not because google feels it is an authority site or that google even gives a rats rump feathers for the page. It is only about links. The only relation it has on rankings is a little added weight given to anchor text in links.

No debate "pagerank is 100% dependant on inbound links" Agree!

If you read my post again, you will see that I am not talking about PR authoritiveness as such, as talking about the authoritiveness that Google places upon the page. Pages that rank highly are those that Google perceive as the authorities for a given term. PageRank has little to do with that as a lower level. PageRank is seen as non-linear as for example: a PR6 page could be said to be 6 times more important that a PR5 page, though a PR7 page is said to be 7 times more important than a PR6 page, not 7 times more important than a PR5 page, thus the higher you go, the more weight is given by Google to that page. We talk about PR4 ranking over PR7 and so forth....shows PR has no resemblance really, however; when you go to PR8 vs. PR9 vs PR10, will normally see the PR9 or PR10 at the top if they are exactly targeting the phrase.

Sorry if I confused you, I am not on about just PageRank here, but think more at the higher end of the scale and how Google places authority to rank pages, ie. 1 > downwards with their obvious PR factors representing links. At the higher end of the market, Googles authority (weight) placed upon a given page, generally easily recognisable by its PageRank, is authoritive in itself to pass authority (in Googles eyes only) to a given page (a vote).

I am not debating how PageRank works, I agree with you in that aspect, I am debating the results at the higher end of the scale and Googles authority it places to those higher end pages containing PR8+. That is a google algorithm factor, not just a PageRank algorithm factor.

[Mel]

IMO there is no direct connection whatever between Googles relevancy rankings of any given page and the PR of that page. If you had an experience where a link from a PR8 page increased rankings fine, but thats not much foundation for a general statement on Googles ranking algo.

I have had opposite experiences like a PR4 page outranking a Google PR9 page for a search engine term. Surely a Google page must have some power as a search engine "authority" but that combined with a PR9 was not enough to beat a well optimized and linked PR4 page. Despite this I am not going to formulate a general ranking theory base on this occurance alone.

[nuclei]

I have to agree with Mel on this. What works one time may not work another, based on different vriteria that were also counted in that one test. Unless there is a decent range of testing done, a formula can not be ascertained just yet.

[Anthony Parsons]

The silly thing is, is that I agree with the way in which the pagerank algorithm works, however; you then see this which completely throws it out the door in every aspect of quantity and not quality.

If the algorithm is correct for instance and works on quantity with a minor variance of anchor text, then this would not happen IMO:

http://www.appliedminds.com which has a PR6, 20 backlinks in total

Pick a site, any site.....

http://www.linkandthink.org which has a PR5, 345 backlinks

or

http://www.justthink.org which has a PR7, 248 backlinks

These are the sort of results that completely throw the original PR calculations out the door and bring in some other algorithm from Google to improve relevancy vs quantity based on the applied PR theories. There are thousands or even millions of these examples that you can look at which do not work within the original guidelines of the algorithm information from Stanford Uni....

This is what creates doubts and makes people like me take different avenues in linking for trial and error which proved the theory wrong also. Because if the algorithm is as per the original document, with nothing other affecting how pagerank is calculated, then these cases would be few and far between and not in the quantities they are. You would not be able to take a PR8/9 site and give another site one link for instant ranking improvement for highly competitive terms and so forth. This just wouldn't work and I totally agree that it shouldn't.....but it does and is.

You could say that this is a penalty against the page linking to something which affects all of the sites that this occurs with, but then why would that affect the PageRank if it is only from link quanitity. Yes, G could change it manually, but not likely. It is either controlled under the algorithms of Google or penalised for a 0 or grey bar. This is what does not add up, regardless how much you read and appreciate the original algorithm formula that is released and public knowledge.

[nuclei]

One minor detail you missed Anthony. Google counts ALL links, not just the ones it shows. Do a linkpop check at marketleap for those sites. :)

And in terms of just PR calculations, you have to remember that in the closest testing, PR goes up in increments of 6's.

[Anthony Parsons]

One minor detail you missed Anthony. Google counts ALL links, not just the ones it shows. Do a linkpop check at marketleap for those sites. :)

And in terms of just PR calculations, you have to remember that in the closest testing, PR goes up in increments of 6's.

Good point, so lets check that.....

The maximum results that one engine picked up are listed below the original figure.

http://www.appliedminds.com which has a PR6, 20 backlinks in total
marketleap results: Total - 20 Google

Pick a site, any site.....

http://www.linkandthink.org which has a PR5, 345 backlinks
marketleap results: Total - 6681 Altavista

or

http://www.justthink.org which has a PR7, 248 backlinks
marketleap results: Total - 875 Altavista

Still not adding up to the applied theory.....

This is why I say that theory and actual (my practical experience) do not make sense, hence why I do not stick with the original theory of the algorithm and post things here that some think are not right, but in fact work. Starting to raise eyebrows now huh????

[nuclei]

Ant, you dont know how many of those links are pr1,2,3 etc

when you add the multiplier it does add up. So many tests have been done that prove the formula. You just do not see all the variables, so it "looks" like it doesn't.

[Anthony Parsons]

Ant, you dont know how many of those links are pr1,2,3 etc

when you add the multiplier it does add up. So many tests have been done that prove the formula. You just do not see all the variables, so it "looks" like it doesn't.

Well that just answered the original question I believe, "is PR Hype"? No it isn't because the PR factor does matter, hence if you link from a PR8/9 page to another the importance factor is much much greater by the formula calculations!

[Anthony Parsons]

Now your thinking a bit deeper Nuclei.....I like your work mate....

[nuclei]

Well that just answered the original question I believe, "is PR Hype"? No it isn't because the PR factor does matter, hence if you link from a PR8/9 page to another the importance factor is much much greater by the formula calculations!


It matters PR wise as to how much PR will be passed to the receiving page. But as was already shown by many here, PR doesnt have much to do with rankings at all if anything. Anchor text carried by the link does.

2 different things. Yet you are trying to tie them together. You can't.

So yes, PR is still hype, when it comes to rankings.

I dont care about PR value of my sales pages. I care about where I fall in the listings.

[Anthony Parsons]

It matters PR wise as to how much PR will be passed to the receiving page. But as was already shown by many here, PR doesnt have much to do with rankings at all if anything. Anchor text carried by the link does.

2 different things. Yet you are trying to tie them together. You can't.

So yes, PR is still hype, when it comes to rankings.

I actually fully agree that PageRank has little to do with rankings....I am not trying to tie them together nuclei....

What I am getting to is this. Quantity of links has no major role in PageRank at all. PageRank is a figure assigned by Google on the relevance and quality of a pages backlinks. Now step sideways of PageRank algorithm onto Google algorithm and you can now assimilate relevance and quality in the measurement of links into other algorithms.

It works in your favour at the higher, very competitive end of the market. Now, when you combine all this with onpage optimization, it works and works well. PageRank is crap, no doubt, but it is still a measurement of an algorithm that is used to calculate rankings.

So in theory, when combined with onpage SEO, if you have a PR4 and PR8 identical optimized pages with one backlink each of the same value, the PR8 should rank over the top in Google because of the link importance factor assigned by Google.

This is probably a bit off topic now from the original topic, but very handy to nut out I think. Many heads are better than one to work through this stuff.

[nuclei]

PageRank is a figure assigned by Google on the relevance and quality of a pages backlinks.

What PR is has been sussed out for years now. And now you are telling us that all the testing, all the evidence, all the proof, is incorrect cause you think pagerank has anything to do with relevance and quality?

For a while there I actually thought you might be heading somewhere intelligent with this.

PageRank is and always has been purely about quantity of different levels of links. It didnt change overnight, and stanford has not changed the pagerank formula to somehow suit someones need to argue that it has anything to do with relevance.

[Anthony Parsons]

And now you are telling us that all the testing, all the evidence, all the proof, is incorrect cause you think pagerank has anything to do with relevance and quality?

Someone once said they would never walk on the moon either? To hard from all the testing, evidence and proof they formulated. So everyone followed, except for a few who developed the technology.

**************************************

Ok, think along these lines then.

Two identical sites targeting the same terms starting from scratch.

We take a PR4 page link and apply it to one, then we apply a PR8 page link to the other. As already discussed, by the theory of PageRank of quantity, both sites have no difference in relevance as yet, however; they do, because one inbound link comes from a PR8 site and the other a PR4.

If the PR of the inbound page had no reflection on PR, then it is solely a quantity job to gain PR and the deciding factor on which site ranks over the other, except an importance factor is slipped in. The importance factor is that Google deems from the two identical sites, that one will rank over the other because it has a more "authoritive (PR8)" link coming too it than the other site which only has a PR4. Take that into the PageRank formula and that is how it works out.

The authority / relevance / quality is what Google is assigning the down stream of sites because of the difference within inbound links PageRank linking into those sites. This authority / relevance / quality is not associated with PageRank directly, but more indirectly how Google places a site within the top rankings. This authority / relevance / quality is not just about links and PageRank, it is about the entire algorithm process, not just links.

PageRank as a seperate algorithm of to the side; Google then incorporates multiple other algorithms to determine authority and relevance upon sites to place them in their rankings accordingly. If two websites in the "golf" industry for example, one has a PR8 and one a PR4. Now the PR8 only has two backlinks both from PR8 like (relevant) sites, the other has multiple inbound links from other sites. Google has deemed through its PageRank algorithm that the PR8 site is more of an authority on Golf than the PR4 in the link aspect only (PageRank Algorithm). On page factors now come into play for assessment. Both pages are identical, except for the links. Again, theoretically, the PR8 should rank over the PR4 if identical because more importance has been ascertained by Google in the PR8's link structure coming from other PR8 (niche authority) sites.

Now, PageRank algorithm is worked out on the side, except that the algorithm is then incorporated into the Google main algorithm. What you may be missing I think within this equation from my explanation, is that the PageRank algo is incorporated. It does its job individually, then is incorporated and measured against other factors. These factors are where importance / relevance / quality / like content / linked pages / etc etc are further measured. So whilst working out the PageRank of a site uniquely from the PageRank algorithm, more is actually going on.

PageRank Technology: PageRank performs an objective measurement of the importance of web pages by solving an equation of more than 500 million variables and 2 billion terms. Instead of counting direct links, PageRank interprets a link from Page A to Page B as a vote for Page B by Page A. PageRank then assesses a page's importance by the number of votes it receives.

PageRank also considers the importance of each page that casts a vote, as votes from some pages are considered to have greater value, thus giving the linked page greater value. Important pages receive a higher PageRank and appear at the top of the search results. Google's technology uses the collective intelligence of the web to determine a page's importance. There is no human involvement or manipulation of results, which is why users have come to trust Google as a source of objective information untainted by paid placement. Source (http://www.google.com.au/corporate/tech.html)

This is where I am talking about relevance, importance, etc. You have to look beyond just the PageRank algorithm to see its importance factors. That algorithm is a unique identity incorporated to be taken further. It only processes one part of the information, not everything to do with links. It is hard to get the exacts from my head onto paper. But it is more than what you are just mentioning. Many are stuck on the pagerank algo as Googles total link factor. There is more to link factors than just the PageRank algo IMO.

Google put it there in writing themselves, PageRank performs an objective measurement of the importance of web pages by solving an equation of more than 500 million variables and 2 billion terms. I don't see these 500 million variables in that one string of PageRank algorithm factor. Lets not forget, this is written under the PageRank heading. Link quality / importance / Page importance the link resides upon / page importance the link points to / relevant terms on both pages / relevant terms within the link text / the list goes on. None of this is mentioned in the PageRank algorithm formula is it? But Google mentions bits and pieces of it their own site under PageRank!!! Interesting isn't it?

[Mel]

PR is not difficult to understand, but trying to understand it by reading Googles public pages makes it difficult IMO, interpreting Googles public statements even harder and trying to make sense out of PageRank by using the Google toolbar numbers really complicates the issue.

It seems clear to me that PR is based only on three factors:

1. Then number of inbound links to a page
2. The PR of the linking page (which IMO is what Google is talking about when they mention importance and/or quality in conjuntion with Pagerank)
3. The number of outbound links on the linking page

Unless you know all three of the above factors for each link on every page in Googles database and you have managed to not only crack the way google represents them on the toolbar but have some way of knowing the exact range on TRUE PR numbers represented by each range this type of comparasion is not possible.

To look at only the simplest aspect of the problem, we can look at trying to determine the number of links google has used in calculating the PR of a Page.

Take one of the sites you have mentioned above http://www.justthink.org

Google link:www.justthink.org search reports it has 248 links, but is only willing to show you 238 of them.

Google search @www.justthink.org reports that it has 1050 results most of which seem to contain links.

This demonstrates that you cannot even estimate the simplest and easiest factor that goes into making up a pages PR.

There is a very good discussion of PageRank at http://www.webworkshop.net/pagerank.html

Now if you try to bring the ToolBar PR into your thinking, the PR reported on the Toolbar has an unknown relationship to the numbers Google uses in its calculations. The numbers Google uses are real numbers, but the numbers reported by the toolbar are tags given to a range of real PR numbers, corresponding in some unknown way way (probably a logarithmic scale) to real PageRank and arranged in a series of ten labelled bins in a way we do not know.

There is no need to try to complicate PR, it is simple and elegant but Google give us neither true PR nor the means to calculate it. When in doubt keep it simple.

As best I can determine the PageRank calculations and database are seperate from the google hitlists and relevancy ranking algorithm, with the relevancy rankings done first and then a PageRank factor applied.

The 500 million variables with two billion terms are the pages and the factors listed above. The little sub "n" you see in the equation means that every linked page in the google index must be included in that calculation.

[steve sardell]

I personally am in agreement with Mel and others. The concept of PR is not difficult to understand, although the actual measurement of individual pages may be. It is simply based on the number and importance of the links. What does get confusing is when the word *quality*is thrown into the mix. PR is not a measure of quality. Quality is totally subjective, and does not fit into a mathematical formula. IMHO, the sooner the word quality is taken away from the discussion of PR the clearer it will be. Sorry Anthony, but importance does not equate to quality, and I believe it is confusing to new readers when they read the word quality interfused with PR.

PageRank also considers the importance of each page that casts a vote, as votes from some pages are considered to have greater value, thus giving the linked page greater value...Google's technology uses the collective intelligence of the web to determine a page's importance

To the initial discussion. Is PR too hyped? IMHO, a declarative yes.

[AussieWebmaster]

Okay Anthony,
That first site:
appliedminds.com
shows no direct backlink on the bottom two backlinks, so what I think has happened is the link it is coming from is actually doing a 301 redirect to the page... as it may be getting from those two or more of them...
any 301 redirect is not a backlink but a permanent redirect passing all PR to the landing page.

[Anthony Parsons]

Sorry Anthony, but importance does not equate to quality, and I believe it is confusing to new readers when they read the word quality interfused with PR.

Yer, I think it is becoming confusing. I will think about what I am trying to portray and start it in a new thread that has no direct relation to PR. I am in agreeance with PR and how it functions, minus a few miscalculations out their. The other factors I am directing too are not the PR algo, but a mere consideration of how that algo is incorporated to mingle, interpret results and produce an outcome that has potential to help many in the highly competitive world. I actually agree with every said about PR. I will take that approach.....thanks.

[Mel]

Hi Anthony
It may help to review the brief explanation of how Google originally derived their search rankings as given in the original Google document, Anatomy of a Large Scale Hypertext Search Engine:

Google maintains much more information about web documents than typical search engines. Every hitlist includes position, font, and capitalization information. Additionally, we factor in hits from anchor text and the PageRank of the document. Combining all of this information into a rank is difficult. We designed our ranking function so that no particular factor can have too much influence. First, consider the simplest case -- a single word query. In order to rank a document with a single word query, Google looks at that document's hit list for that word. Google considers each hit to be one of several different types (title, anchor, URL, plain text large font, plain text small font, ...), each of which has its own type-weight. The type-weights make up a vector indexed by type. Google counts the number of hits of each type in the hit list. Then every count is converted into a count-weight. Count-weights increase linearly with counts at first but quickly taper off so that more than a certain count will not help. We take the dot product of the vector of count-weights with the vector of type-weights to compute an IR score for the document. Finally, the IR score is combined with PageRank to give a final rank to the document. (emphasis mine.)

While this is an old document I believe there are things to be learned from it, one of which is that the original design was to calculate a PR rank for every page in thier index, and store that in a seperate database, then create the hitlists that contained the relevancy data and store those in different databases. The relevancy calculations were first computed as outlined above and only then was the PR brought into the final rankings. It is noted elsewhere in this document that anchor text is stored in still a different data base and then brought into the hitlists.

This approach allows different machines to record data and perform different calculations, with much of the processing done in advance, and then search only a subset of that data to produce the final relevancy rankings, which are then combined with the precomputed PR to provide a final score.

Based on the above I believe that it is useful not to assume that PR has anything to do with relevancy rankings.

[steve sardell]

This approach allows different machines to record data and perform different calculations, with much of the processing done in advance, and then search only a subset of that data to produce the final relevancy rankings, which are then combined with the precomputed PR to provide a final score.
Based on the above I believe that it is useful not to assume that PR has anything to do with relevancy rankings.

Mel, I totally agree with your posts. Because of time restraints these calculations need to be done in advance. It would be grand if all could be done on the fly, but both the hardware and software are not yet there. I believe Teoma has been attempting, but the best time they have been able to produce is barely under a minute. Way too long to wait for search results.

Anthony, I think you are merely ahead of the curve, and there is nothing wrong with progressive thinking; it is a plus. IMHO G in particular will eventually include some type of relevance factor into the PR measurement, however, they are not quite ready.

[nuclei]

If your niche relates to that particular niche with a higher PageRank, then it does share some effort towards demonstrating the authority of the site your link is being placed upon that will pass some authority too your site (this all being related niche content).


Ant, PageRank does not give a damn about niche or theme at all. It has *nothing* to do with quality or relevance in the least.