Does PageRank really matter for ranking in Google?
August 9, 2007
PageRank has been all the rage since Google first introduced that little green line on the Google toolbar. Links are bought and sold at a premium based on how far that little green bar stretches to the right.
But is it really wise to pay a premium for links on high PageRank pages? Are sites with the highest PageRank the ones that are ranking better in Google?
I decided to test that theory, so I ran 500 very competitive keywords through Google and checked the PageRank of each of the top 10 ranking sites for each set of keywords.
Here's the average PageRank for the top 10 ranking sites across the board:
1. 6.722
2. 6.866
3. 6.292
4. 6.234
5. 5.968
6. 5.88
7. 5.73
8. 5.662
9. 5.656
10. 5.604
Now, just looking at the averages, you might say, "wow, I guess high PageRank sites do rank better!" And you wouldn't be entirely wrong for saying that. Yes, overall, sites with higher PageRank are ranking better in Google.
But the question is: is it the PageRank that's causing the high ranking, or is it something else? I decided to dig a little further and find out.
Here is a list of the number of times for each ranking position that the ranking site has a lower PageRank than the site beneath it in the results. For example, if a site with PageRank 5 ranks #3 for a set of keywords, and the site ranking #4 has a PageRank higher than 5, that adds one to #3 in this list.
1. 155
2. 133
3. 148
4. 152
5. 167
6. 178
7. 165
8. 187
9. 178
So about one-third of the time sites are outranking other sites with higher PageRank, demonstrating that higher PageRank doesn't always win.
Here is a list of the number of times for each ranking position that the ranking site has a PageRank at least 3 points lower than the site beneath it in the results. For example, if a site with PageRank 4 ranks #3 for a set of keywords, and the site ranking #4 has a PageRank of 7 or higher, that adds one to #3 in this list.
1. 85
2. 64
3. 68
4. 55
5. 64
6. 65
7. 60
8. 72
9. 62
So (roughly) 14% of the time sites are outranking other sites that have a PageRank a full 3 points higher than theirs. Google must be looking at something else to rank these sites.
Let's take a look at a few specific example of this happening, then I'll talk about why this is possible.
As of this writing, the site that ranks #7 on Google for the phrase "play games" is playrelax.com. This site has a PageRank of 4. Care to guess what ranks #8 for the phrase? Yahoo! Games — a site with a PageRank of 8!
Let's take a very competitive phrase as another example. As of this writing, the site that ranks #4 on Google for the phrase "airline tickets" is traveldiscounters.com. This site has a PageRank of 5. The site that ranks #5 is travel.aol.com, with a PageRank of 7.
One more example, just for fun. As of this writing, the site that ranks #1 on Google for the phrase "newspaper obituaries" is newspaperobituaries.net. This site has a PageRank of 4. The site that ranks #2 is legacy.com, with a PageRank of 7!
Are you stunned? Does it seem amazing to you that a site with a PageRank of 4 can beat out a PR7, or even a PR8 site, ranking ahead of the much higher PR sites in Google?
Don't be stunned. It's not really amazing.
To understand why this happens, you need to understand how a site gets PageRank. A high PageRank can be achieved two ways:
- Getting a lot of links from low PageRank pages.
- Getting a few links from high PageRank pages.
So if you got a million links from low PR pages, you're going to have a high PageRank, or if you get a few links from very high PR pages, you're going to have a high PageRank.
But you see, it's not the PageRank of a site or page that Google focuses on primarily when ranking a site for a set of keywords. It's the keywords that those links contain that is of the greatest value in Google's algorithm.
For example, the PR4 site (playrelax.com) has 1,231 links (according to Yahoo!), whereas Yahoo! Games has a boatload of links (1,212,619). How is it possible that playrelax.com is winning the game if it has so few links in comparison?
I can gurantee you that most of the links pointing to Yahoo! don't contain the keywords "play games" as often as they do in the links pointing to playrelax.com (and many of Yahoo! Games links come from their own site). But the owner of the other site, playrelax.com, probably realizes that in order to out-rank the others, his site must have links that contain the right keywords. He gets the right kind of links, and he wins the game.
No doubt this is the case with almost all of the lower PR sites that are beating out their higher PR competition.
Now, generally speaking, the sites with higher PageRank are focusing on the keywords in their links, which is why (on average) higher PageRank sites are holding the top positions. But this is not always the case. As I said, about one out of three ranking sites have a lower PageRank than the sites they are out-ranking.
So does PageRank matter? No, not really. What matters is that you get quality links from other sites that contain the keywords you want to rank for. Do that, and you might just find your site outranking the big boys, too.
I and hundreds of others are using 3WayLinks.net to get the right kind of links and rank our sites, and at a price that won't break the bank!
Please post your thoughts on this below.
Comments
64 Responses to “Does PageRank really matter for ranking in Google?”
-
Jonathan Leger has been earning his living online full-time since the fall of 2004. He regularly posts advice about search engine optimization (his current passion), AdSense and Internet Marketing. 
-
Hi Jonathan,
Some interesting analysis. I agree with your conclusions. By way of example I started a particular blog a few months ago which has yet to achieve a page rank. Using 3WayLinks.net I have managed to achieve position 1 in Google for a pretty competitive phrase, going from position 500+ in less than a month.
Keep up the good work
Chris
Thanks Jon for explaining PR positioning. This has been a mystery. I wonder if keyword rich articles also boot a site to high PR too?
Keep it up.
Jim
Jonathan
Your testing is on track. As you stated, we don't exactly know what results in the current Serp-rankings. But we CAN look for patterns to follow. Good anchor text in backlinks is likely an important factor. Good Sleuthing today!
Ok, another example:
I did some work for a guy: www.theoakflooring.co.uk (PR1 !!!)
Do a search for "oak flooring", and (In the UK at least) this site comes up at #4.
The site at #5 has a PR of 5.
If you look at the other sites on the list, you will find that this site is in fact the LOWEST PAGE RANK on the results page, with the rest being 2-5.
Interestingly, the one above this site has a PR of 2.
It just shows this: Superman the comic hero has a lot of raw power. That doesn't mean that he is any good at knitting.
My mum doesn't have a lot of raw power, but I know she can knit.
Page rank is a basic measurement of your raw power, how you channel it is up to you.
By the way, I am trying to get php enabled on this domain to get 3waylinks on the site so I can branch out to other keywords easily.
Cheers.
I don't think I would base my conclusions on a few exceptions.
Page rank is pretty much worthless on predicting link value and rankings. They could have a pr10 and still not rank worth a crap because maybe the on page optimization isn't their.
When link building think of brand awareness, direct traffic, usefullness, who they link to….but do not get caught up in the Page rank…It's been gone for quite some time now…well atleast I thought most knew that except for newb's
Kevin:
33% doesn't qualify as "a few exceptions."
James:
You would think people would realize that, but they're still paying big bucks for high PR links.
Oh also we could just buy a domain say "coolest blah blah blah" and hope it helps in rankings. Which it will! So don't waste time on buying links for page rank because I feel it will not help…atleast not much!
Everyone Jon is right page rank is gone with the wind…just like supps..Page rank means nothing anymore. Period. So while your itching and waiting for the next page rank update and hoping it moves up then just know Google really doesn't use page rank as a factor for ranking. Trust me I have seen it to many times.
Hi,
I like the last example with the "oak flooring".
It demonstrates (to me) very clearly how important it is to have not only relevant content and keywords but also headlines and "domain names" according to your theme / niche!
In this example theoakflooring within the domain name does the trick.
Just check the previous example again.
Hi Jonathan,
That is some great information for those who are just starting out. It can seem like an impossible task for a new site to get a good top position, but as you pointed out, you've just got to do it the smart way. What do you think the difference in traffic is from a top 5 position and a position on page 2 or 3?
Ted
Very nice observation and analysis. I agree in your points in the aricle, and that high PR not always ranks you best. The anchortext of your backlinks need to be correct in order to achieve a higher ranking. What's interesting is the Yahoo example which clearly explains the theory of high PR not always is ranking highest. Even though I think high PR with the right anchortext is the most succesfull solution, but it is also the most expensive. Therefore low PR links with the right anchortext in i large scale would do almost as good, if not better I think.
Hi Jonathan,
Good analysis. This is one of the reasons I strongly suggested that 3WL needs to have custom anchor text in the outgoing links to the "partner" sites. Even better would be the ability for the domain owner to customize each anchor text coming in from the inbound partner site links, so that we can choose a range of longtail and shorttail keywords for our sites.
Cheers,
Peter
PS - no sure if shorttail is a word yet
Peter:
3WL allows for 3 sets of link text to be used, in a 60%, 30%, 10% split (which I've found to be most effective).
3WL is not a one-step solution for all ranking. I always recommend writing and distributing articles as well, to get more inner-page links. 3WL is meant to give you a strong foundation for ranking.
3WL is powerful enough that it is, indeed, ranking sites on its own merit for mid-level keywords, but for maximum benefit you should always work to get more links to your site's inner pages.
Hi Jonathan,
I liked that bit of research into it and it proves out something else that was covered recently at an internet marketing seminar that I attended.
The main speaker (I really hate having incomplete information, but I can't remember precise names right now) who we'll call Alex had stated that the PageRank bar really had nothing to do with your actual Serps ranking.
He had stated that it was actually a method for gauging the authority of a site, based on it's relevancy. But it doesn't do much but gauge links, which are then weighed for relevancy to the site in question.
So it's like a gauge for equivalent relevancy, rather than direct relevancy. This is why it comes up differently in the SE results, and doesn't consistently match with the PR of a particular page.
Alex also pointed something else out. It was named for the guy that created the PR system, whose first name I can't remember, but it's not relevant. What is relevant, is his name was Mr. Page… Hence, Page Rank.
Now, while he did give a first name, I didn't follow up, to discover the truth behind this, but they also showed a picture of the fellow, so I'm guessing it was accurate information.
So Page Rank doesn't have anything to do with "web-page ranking" directly, it does have to do with "link popularity ranking via authority sites" which has little (or at least, only a small bit, as in one factor) in Googles determining the particular ranking relevancy for a keyword.
This, of course, is only according to my particular understanding of it, and I'm no SEO guru.
Peace & Prosperity,
Jamie Davidson
All this talk of backlinks begs the question as to why you have 'no follow' set as default on this blog Jonathan? If you moderate your comments why not share some link love?
Great stats and information Jonathan. You really explained it well.
Keep up the great work!
Gary Knuckles
many webmasters want the pagerank tool removed from googles toolbar because of the misconceptions and link buying it generates
check out this thread at webmasterworld-
http://www.webmasterworld.com/google/3416514.htm
google rep actually replies and states that all suggestions are welcome
I've got the no follow disabled in WordPress Clive. I'm not sure why it's still putting them. I'll have to contact semiologic support. Thanks for noting that.
Hey Johnathon,
Great post. I would definitely have to say the page rank isn't what it use to be. I'm not sure what it actually means for sites today.
One great example is one of my sites has NO page rank and it comes up for almost all my main keywords on the second page of google.
Elizabeth
Excellent research, Jon! Once again, you've outdone yourself with your in-depth research.
Thanks for creating 3WayLinks because my website has already jumped another rank closer to the #1 position since joining 3WL. What's amazing is that I've only been a member for a little over a month.
Keep up the great work.
Something weird happened with one of my site PR after 3waylinks.net. It dropped from PR3 to PR1. I did not know what happened.
I almost got upset.
But then the rank of that site for certain keyword was astounding. It jumped from page 6 to page 2 of SERP out of 1.7 billion search results!
Isn't that amazing?
Forget about PR.
Hi Jonathan,
Nice article. If I'm on the top 5 that's great and anything less is unacceptable. I use LSI.
Thanks,
Rusty
That was a great article. One of the best that I have read that really explains how this darn page rank thing works. Great stuff, keep up the good work.
I just wanted to see if this links back to my site. I really enjoyed the article, truly amazing. I am outranking some sites with higher page rank. I am using seo elite to help me win this game. Thanks again.
I have to agree only 10%. The PR you see is just a toy not the real google PR which is updated every day.
3 way links are easy to break thus leaving you with one way links and no effective way to check that it is a complete link. There are quite alot of other factors to consider that you dont even mention. In my targeted key phrases on google I place the top 4 with 2 different companies ad a pr of 2 with just 4 links. That is a SEM phrase too.
They use to say that content is king, now they are saying unique content is king. Has anyone ever considered that the % of content to code could be king? I can put any website at the #1 position as long as I build it right.
Thanks
Don Young
Don:
Even though yes, the toolbar PR is not the real PR, I doubt seriously that Google is showing a PR4 on the toolbar for a site that is PR7+ (and still these PR4 sites are outranking the PR7/PR8 sites).
3waylinks.net automates the verification of links, and only allows quality sites into the network. That's why I created it — because it is so difficult to handle manually.
Good going on your rankings. Once you know how it's done, the world, as they say, is your oyster.
Jon
I am not sure that explanation works.
Search for twain harte rental in Google.
Here are the links for various sites:
#1. http://www.twainharterentals.com/ - links: 90
#2. http://www.thvr.com/ - Links: 109
#9. http://www.funcabinrentals.com/ - Links: 622
Now most of the links to #9 use the keyword as the title/link text and the other's don't… Any ideas why the other sites rank so much higher?
I agree.
I have examples of 0 PR website overranking PR3 and PR4 websites. It's not about PR all the time. Maybe not at all.
PR means a number of Backlinks to a specific website. If a website have for example 100 incoming links probably few of them (lets say 10%) are with good anchor links. Meanwhile, a website with 10 incoming links will probably have less "important" links pointing to him. Just because it has lower number of links.
But if this website with only 10 links get good linking - he can defanetly get a higher Rank.
Hi Jonathan,
apart the 3waylinks network (of which I'm a member) which ways exists to get incoming links with good anchor text ?
Thanks in advance
alexander
Jon,
Thanks for your analysis it is as always appreciated, I agree with your general conclusion but for a different reason.
The fact that a page has a Google pagerank of 5,6,7,8 etc is irrelevant in a specific search term - they may for example have gained the page rank as a result of lots of links from high PR sites with the link text of "win games". However if you use the search term "play games" they may not have been optimising for it - hence they can have a high PR but rank lower for a given term…it may not be their key term.
In addition…I am also on 3 Way links, this is the best sneaky affiliate post I spotted for a while
Dave
Hi Jonathan,
Wow — this is fantastic information to put to use for those that are just getting started or those of us who haven't a clue!
I'll make sure to bookmark this page & keep my eyes open for more replies on this subject.
All The Best!
Cathy Lee
Interesting post, as usual, Jon. Well done.
I have noticed that my PR4 site does not always do as well as other sites with lower PR, and you have helped to explain a lot of why this is so. Fortunately, in most cases, my lower positions are for keywords that I am not targeting, but it is still nice to have high search engine positions.
I was under the impression that all my positions would improve because of my page rank - bummer! : - )
Ok everyone, really what does everyone look at when they are looking for page rank…Link value right.."I want my link on their page rank 7 site so I can increase my rankings for "Blah blah".
The thing is, everyone is targeting high search volume, low converting keyword phrases. What if you could target high converting kw's and push them to the top of Google as much as you wanted? Then what would happen? Your sales would increase right? What if I told you in my niche I dominate the serps for long tail, high converting keyword phrases as much as 8 times in the top 20. And you know what almost everyone is ass backwards when it comes to SEM. People target keyword phrases that result in nothing but browsers not buyers. Why is this? Why target a kw that gets searched 100,000 times and converts at only 1% instead of targeting all of these long tail keywords that convert as much as 30 even 40%. Little streams make a huge freaking river.
Once again what if I told you I dominate the serps as much as 8 times out of top 20 for high converting keywords and I can do this as much as I want. So if you want sales then you shouldn't be targeting high search volume, high competition, low converting keywords.
So forget about PR and focus on what matter sales. And if you are not getting sales then you are not leveraging web 2.0 to your benefit!
I agree with your conclusions. As an example, I started a particular site last month which has yet to achieve a page rank. Using 3WayLinks.net I have managed to achieve position 1 in Google for all my 3 competitive phrases, going from nowhere to No. 1 in less than a month.
nice post Johnathon -And lots of feedback. The only possible bonus of PR is that the spider may come back more often if its a High PR- I have noticed that more focused anchor link text links helps. I always like the facts that back things up……
Thanks Jonathan for this very informative analysis.
I can really use this analysis to inform my students of my current project "wordpress Blogging Coaching Program" here in Malaysia. My partner shared with me on how he had always wanted to rank well on natural search engine Google on keyword phrase "small business marketing" for some time.
After he used 3waylinks, his blog went up to page 1 on Google natural search. So it looks like I will join in the ranks here with you and others.
the research is interesting, however, from a small set of data, the conclusion is not so convincing.
google search results depends on 2 main things:
1) links;
2) content the site has;
PR is calculated based on inlink and outlinks. but it is not the only thing the search rank uses. although 1) is important, 2) may have serious impact on search results.
recommend a book:
Google's PageRank and Beyon: the science of search engine rankings.
I'm interesting in 3WayLinks.net but a question is can I sign up with sites that running in sub-folder.
Good research and an eye opener for all website owners.
Good luck.
Hi John, question for you…
I'm a paying member of 3waylink and I asked the Help Desk why the sites linking to me weren't relevant (sites similar to mine).
The Help Desk's response via you was that
there haven't been any studies proving
that you need the sites linking to you
to be relevant to get good ranking.
In your blog post, it seems that the site
that ranks well for the terms "play games" does so because it has more RELEVANT sites linking back to it, per what you wrote.
So what's the deal here?
Do we need relevant
sites linking to us or not?
I'm getting mixed signals here…
Maybe I just don't understand
enough about this, perhaps?
Thanks,
Dan
Daniel,
If it is a one way link it most certainly does not need to be relevant. Why, because anyone can place a one way link to any site they choose. Also one way linking is one of the most powerful links you can obtain, outside of a few other methods I use.
Now reciprocal linking I feel is more powerful if they are theme related and compliment each other. However they not need be relevant it's just the power you are looking to get out of the link. Theme related more powerful, non related less powerful.
Hi Jonathan,
I have been your reader for almost a year, you are really an incredible entrepreneur.
Excellent post and excellent blog. the keywords and pagerank combination is the key to success. Thanks for share this invaluable information.
Best Regards.
Daniel and James:
The links do not have to come from relevant pages, but need to contain relevant anchor text. My studies and my own experiences are proving that sites can be ranked well even if all of their links come from off-topic sites.
That's not to say that on-topic links aren't better (they may be — though I've not seen this studied and presented with real facts). What's bogus is people who say that links coming from off-topic (or off-theme) sites are ignored or not counted by Google. That's bunk, as my own post on the matter demonstrates.
Jon
Craig:
Regarding "twain harte rental". I haven't looked at it in detail, but Google also uses what it calls "TrustRank" (a figure it does not publish publicly) to rate the "authority" of one site over another. Getting links from sites Google has deemed more "trustworthy" can help your rankings, too.
I'm not sure that's the case for your example, but it's one possibility.
Jon
Alexander:
To get more links, write articles and distribute them via EzineArticles.com, GoArticles.com and ArticleCity.com.
Jon
I have a concern about 3 way links. Maybe you can give me an answer. As Google gets more sophisticated won't they be able to recognize that you are doing 3 way links? When they do "catch on" might they not consider those links spam?
John,
You are right about theme links. What I said was I do not believe non related links are as powerful as theme related links. Also the whole trust rank is very real. If you get links from authority sites then yes they will be a lot more powerful than your run of the mill link.
Also who you link to will also play a part in your rankings. Try throwing a link on your home page to Google, wiki or an authority site…Then see what happens.
John,
I think you are wrong about receiving links from Goarticles.com because for some reason I have submitted several articles their over a long period of time and you know what none of them are showing up as links in the webmaster center in Google. However every other article I have submitted else where is showing up.
Wonder if you can look at that to see if its just me
Thanks
Dee:
I say "no", because I am aware of a number of 3 way link networks that have been operating for years. Think about it: if the sites in the network have links coming in from other places, and going out to other places (in addition to the 3 way links), how is that any different from "natural" linking?
Reciprocal links are easy to spot, because each site links back to the other. Three way links are much more natural in their appearance.
Jon
James:
I'm talking about getting links from the people who pick up the article from GoArticles.com and post them to their own sites — not the links from GoArticles itself.
It would be wonderful if everyone could even focus on page rank, but for the average internet marketer of affilate programs who may also maintain a blog or two… it is excellent that google searches bring up your information based on something as simple as BethsVentures Adlandpro !! Try it !!
The fact that anchor text counts a lot brings back the discussion on how to choose the keywords wisely, regarding this I have a couple of questions.
If I want to rank good for "make money" and for "passive income" is it wise to choose the keywords "make money with passive income" ?
Will I in this way rank well also for the keywords "make money" and for "passive income" or only for the whole exact phrase "make money with passive income" ?
Thanks in advance
ciao
alexander
John,
OK, gotcha the publishers. Yeh but I think its kind of weird that the links directly from goarticles do not show up.
John,
Got a question maybe you can help me out on this one. What do you think differs from Google U.S. algo compared to say UK, Ireland, Canada and other countries? Besides domain extensions,indexed pages..
Thanks John any thoughts would help.
John,
PR & quality anchor text is two thirds of the equation, the missing one third being how closely the content of a site linking to your site matches the keywords in the anchor text. This is how Google and others qualify the 'relevance' of an otherwise good link to your site.
If the theme of a site or a page that links to your site doesn't come close to the keywords in the anchor text in the link to your site, it's not going to help either your PR or relevancy a lot.
One of my sites is PR4, but only has some 140 external sites linking to it. It's all about relevancy, not about sheer numbers.
Awesome, John.
This was an awesome experience. As I always say, PR doesn't bring money in the bank… Unless you are actually flipping websites!
Hold on Jon, don't try to figure out or reverse engineer the search engines if you want to keep your sanity.
I out rank Wiki for the search phrase - article assistants - and I have a page rank of -0- !!
I am #2 in msn for - how to make homemade wine- with a page rank of -0- !!
I have a blog that is #23 in Google after only 5 days, obviously no page rank yet.
The thing I notice is the relevance of results in title, description and domain.
The most relevant results for a search phrase comes from the amount of times that search phrase is found in the title, description, domain and meta tag keywords.
For example:
Article Writing Tips, Professional, Standard and AssistantsArticle Writing Information, Software Sugestions, Helpful Products Related To Writing An Article And Content Creation For Blogs, Websites, White Papers, …
{snip: no personal urls please} —this is the result in Google for the term article assistant where I out rank Wiki.
This is the result for Wiki: Professor - Wikipedia, the free encyclopedia Please improve this article or discuss the issue on the talk page. ….. The positions of Reader, Assistant Professors, Associate Professor and Professor in …
en.wikipedia.org/wiki/Professor
My title, description, domain and meta tags contain the phrase more times than Pr3 to 7 sites therefore my site is perceived to be more relevant than theirs.
When setting up a website, it is important to optimize the site for search results by incorporating your phrase or phrases in these areas.
My blog at {snip: no personal urls please} was #23 in Google results in 5 days for the phrase -wine making instructions-because of my optimization methods….not for content because I had only made one post.
It's easy to get in the first page of the S.E.s for a phrase, the most important part is providing great content that will convert those newly found visitors into customers, subscribers and return visitors.
Thanks for the research, most of us don't have the time to research in such depth.
Brian Ankner
Dear Jon,
How are you. I hope you are doing fine. I read you eBook where you have very clearly explained how all things work by using your website for the keyworkds "cat picture" I learned a great deal from you.
Thank you!
Aftab
Never thought about that! I always thought that Page Rank increases the reputation of a website but does not significantly improve a website's ranking. Thanks for another good post.
There is one problem you are always going to have with this analysis. Any analysis you do is based on comparing old information to a current unknown situation. Any PR you see is not what Google has on record at that time.
Currently rankings on the same data centre can change several times throughout the day. So you are constantly trying to understand a moving target.
Have fun trying to understand the method Google use but don't waste too much time on it…
Just some thoughts on links:
1 way links are good.
one way links from related sites are better
one way links with relevant keywords in anchor text from related sites are even better
one way links with relevant keywords in anchor text from well established related sites are best (these would be sites with age, tons of content, and tons of relevant incoming links themselves)
Do you need to get these? No. But why you wouldn't try to focus on them instead of just random incoming links is something I'm not sure I understand.
Maybe I misread some of the posts, but it seemed like there was a lot of talk along the vein of "just get links".
Just some thoughts.
Cheers.
If you have extra money to burn, then you can go on buying high PR links. Your high PR won't matter on search result, but in my opinion (based on my own experience) links from other high PR sites matter most on the ranking (given that the anchor texts match the search keywords).