JLH Design

Much has been said about Google’s love affair with the wikipedia and I am going to say some more.

The wikipedia apologists will point to the fact that they have garnered so many natural links that they deserve the rankings. They will also point out their extensive interlinking that helps the site boost its own rankings based on it’s own authority. This concept is foreign to me but it must be true in the world of Google*.

Imagine if you will, a new way of analyzing data developed by two bright college students in their dorm room. Their way was so revolutionary that the dream quickly (in corporate time) grew to be a multi-billion dollar company touching the lives and pocket books of everyone involved with the internet. This data analysis method IS their brand, it’s what separated them from the rest of the also rans in the great technology race. They invented it, they own the rights to it, only they know exactly how it works. Before they wrote about it, it didn’t exist. You get my point, I hope, they are THE authority on the subject. In their words it is, “The heart of our software”

Fast forward to today. If you are interested in learning about this concept which built arguably the web’s strongest performing property you would probably Google the trademarked name: PageRank.

Who would you expect to be #1 for that query? The founders and inventors of the term? The corporate web site of the trademarked name with possibly millions of links to it? I would. Apparently however, Google thinks that the wikipedia is more of an authority on what PageRank is than Google itself.

If this isn’t a serious indictment of Google’s unnatural propensity to return the wikipedia in their results, I don’t know what is.

The one thing wikipedia is good for is references of benign facts and a cursory overview, definitely not an authority on the subject, so using them for what they are good for I found this citation:

Many years ago, there lived an emperor who was quite an average fairy tale ruler, with one exception: he cared much about his clothes. One day he heard from two swindlers named Guido and Luigi Farabutto that they could make the finest suit of clothes from the most beautiful cloth. This cloth, they said, also had the special capability that it was invisible to anyone who was either stupid or not fit for his position.

Being a bit nervous about whether he himself would be able to see the cloth, the emperor first sent two of his trusted men to see it. Of course, neither would admit that they could not see the cloth and so praised it. All the townspeople had also heard of the cloth and were interested to learn how stupid their neighbors were.

The emperor then allowed himself to be dressed in the clothes for a procession through town, never admitting that he was too unfit and stupid to see what he was wearing. He was afraid that the other people would think that he was stupid.

Of course, all the townspeople wildly praised the magnificent clothes of the emperor, afraid to admit that they could not see them, until a small child said:

“But he has nothing on!”

This was whispered from person to person until everyone in the crowd was shouting that the emperor had nothing on. The emperor heard it and felt that they were correct, but held his head high and finished the procession.

Denial is not just a river in Egypt, it’s time they realize that the search results are LESS authoritative with wikipedia at the top.

---

*It reminds me of a class I had sophomore year in college. I took an elective called, “The Philosophy of God.” Don’t let the name fool you, it wasn’t a religious class but rather an exploration of the human psyche’s need for defining a being higher than themselves. One afternoon’s discussion lead to the concept of the unmoved mover by Aristotle which we had to try to reconcile with the big bang. When asked to list possible problems with the big bang I of course fell back on my physics training and cited Newton’s 3rd law ( for every action there is an equal and opposite reaction) . When pressed further by the never shaved, rarely bathed, and haphazardly dressed doctorate of philosophy for a less engineering based illustration I challenged him to reach around, grab his pants around the waist, and lift himself up and fly around the room. Being the odd fella that he was, he attempted it to the amusement of the class and to further prove my point. That moment earned me my philosophical handshake, an odd moment when the professor got on bent knee at the head of the class and extended his hand for a handshake. They were given out for simple and brief explanations to complex situations. This story would not earn me such a handshake.

I’ve seen a lot of seemingly good sites loosing their PageRank completely. It’s propagating through the datacenters like a normal update. Perhaps there is an update coming? Or perhaps, just perhaps, Google has finally disbanded the entire visible PageRank system and we can all get on with our lives after inevitable the discussion dies down. I can only hope.

If they are not disbanding it completely, hopefully they are going to move inside Webmaster Tools along with the link command results.  (fingers crossed)

Think of the implications!

• The link selling industry would loose its most abused form of assigning value to a link.
• Link seekers would be forced to judge a link based on the traffic and type of traffic a page produces rather just the link juice.
• Digital Point and Webmaster World won’t have anything to talk about.
• SEO’s and webmasters will have to buy Camaros or Porshes to make up for the lack of PageRank inspired phallic replacement.
• Alexa Ratings, the most easily gamed metric ever developed, will have see increased use. I’m sure the Amazon people are salivating at this very prospect.
• People will start publishing their site stats, feed subscriptions, and other metrics to try to justify their existence.
• Site owners will worry more about search engine rankings and less about Page Rankings.
• Google will be admitting that even though PageRank was a great idea as far as indexing goes, the concept of making it public was a mistake.

I’m sure there are many more, any one else care to elaborate?

An interesting dialog took place on the Google Webmaster Group, since we can’t get any credit for it, and a finite amount of people read it I thought I’d transpose it here to save it from falling out of site on the group.

The conversation is as follows.

Admin Aaron

I am wondering is PR is finite? Websites that do not do link building and obtain backlinks in a natural organic way often have very few incoming links which means lower pagerank.

I have a few blogs in the renewable energy area where I find, review and write about companies, people and technology I believe will encourage the growth of alternative energy.

I am generous and link out to others but sinse the site is new and relatively unknown I am not yet getting many links to bring in more pagerank. In other words I am giving more PR out than I am getting.

The question:

I recently started using the nofollow tag on external links with the belief that pagerank is finite and I only have so much to share. Am I shooting myself or those that I link to in the foot? Is this an incorrect assessment of how to pass PR?

I believe this to be a highly relevant question and example of why nofollow tag needs even further explanation.

Thank you if you feel comfortable with answering this Adam Lasnik or others…

Aaron

Sebastian

Aaron, you shoot yourself in both feet. PageRank hoarding is a sin. PR is a sexy fay you should not pass, that’s way to rough. Let her flood your site, let her come and go as she decides, don’t even think of her when you deploy links. And please remove the f** nofollow values in your rel attributes, that’s unethical and counter productive. Really
Here is more info:

Sebastian

Admin Aaron

So if I have a new blog with pretty much no PR is makes sense to link out to all my friends? I don’t know Sebastian, in an ideal search engine world that would be great but I am not so sure pagerank in infinite. I have also been thinking the way you do for a long time but now I am not so sure man.

The way I have been using the nofollow is if I write about a grat subject on another site but that site is filled with paid links I use the nofollow because I do not want to pass PR to a bunch of spam sites selling links BUT this recent idea is driving me nuts. ;-(

Admin Aaron

Yo Google, please enable an edit feature for when we spell stuff wrong we can go back and fix it. Just allow people to edit within a given time period… like 5 minutes? Also how about a spell checker for us dumb arses? Thanks! =P

JLH

Aaron you bring up a good point. A while back when some sort of penalty was being dolled out Adam repeatedly said here and other forums that one factor may be “Over Optimization.” To me a site that has all external links nofollowed is ripe to be picked in an algorithim for over optimizations because of the un-naturalness of it. It’s a strong signal of two things 1) this guy doesn’t trust anyone he’s linking to or 2) this guy is trying to hoard page rank. Neither is a signal that should help a site rank for anything.

Back in the day search engines ranked based on the text only, keyword counts etc. People started repeating “free porn” 900 times on a page, then they started looking at the META data people started stuffing them, links were good so people started selling them, in all cases google reacted and penalized the offending sites. Nofollow was introduced as a tool to help eliminate the value that blog comments, message boards, forums, etc where having on the ranking of sites. If you’re the site owner and didn’t write the content you have a way of saying I cannot vouch for what’s there. I would bet that if someone
tries to turn that around and use it to IMPROVE their page rank, they’ll get busted. Just as using your keywords in sentences is good but repeating them 100 times and abusing the H tags is bad.

Another thing to consider is from the inherent nature of the “web” sites are not meant to be islands but rather connected to the whole internet by both incoming and outgoing links. A site is judged not only by the content and anchor text of incoming links but the topical nature of the sites linking in and the topical sites that you link to. If you have all nofollow links you are missing a key ingredient to telling Google what the site is about.

I don’t have any insider knowledge, but given enough time patterns will emerge in the mass of data that they pour over every day, and I wouldn’t doubt a signal they will look at is your linking habbits and the use/abuse of the nofollow tag. PR hoarders will be penalized as its a breakdown in the natural flow of the web. It’s like the wiki debate, if they don’t trust any of their resources then they shouldnt be trusted as one themselves. Of course they’ve got enough momentum that their probably impervious to any sort of algo change, but smaller sites are.

So where is the line between SEO and SEOO (search engine over- optimization)? I’m not sure, but as Adam also says, “does it pass the smell test?” And saying, “I changed all my links to nofollow so I don’t leak page rank” smells like you are trying to game the system and artificially influence googles ranking, which in my experience they frown upon and react.

Personally, I always have nofollow links highlighted in my browser with a bright red box. If I stumble upon a site that has too many, I move on, I wouldn’t doubt google takes that stance sometime in the future. I just don’t trust the nofollow actions yet, and I think its going to get worse. If you really want a bunch of sites that you link to be not followed for ranking purposes (affiliate links, selling links for traffic only, etc) I’d put those on a redirection hidden behind the robots.txt. That way Google isn’t going to follow them at
all because they aren’t going to see them, and they are not going to be able to hold it against you because there is no possible good that can come from a page that is not crawlable. As far as a site with no eternal links, I’d say that’s just as shady as disabling the back button, the site is no longer a part of the web, but only a town where all the streets going to it are one way, and after a while people will notice that no one returns from that town, and no one will go there
anymore. Now I doubt Adam will/can come on here and say that they penalize or don’t penalize or give credit for the use of nofollow as commenting on
the actual ranking is probably not allowed in the least bit, but these are my two cents

Sebastian

In the case of linking to a site plastered with affiliate links and ads, where is the point to write about them in the first place, if you don’t like them? Well, given that this site has a page hosting valuable information, wouldn’t it be the search engine’s job to allow or disallow incoming PageRank at this site? Why should you pre-qualify links for Google’s ranking algo? This attitude smells like self-
censorship based on fear, and you shouldn’t fear Google when you link out.

Applying nofollow crap to affiliate links is a completely other story. I have no problem with Google’s position that affiliate links should not carry Googlejuice, but just the human traffic. I say “nofollow crap” because rel=nofollow is IMHO not the suitable instrument to achieve this particular effect, or call it favour to the engines. I use rel=nofollow with affiliate links to save my own ass, although
it’s Google’s job to decide whether a particuilar link should carry weight or not, and although I do think that rel=nofollow is not the right thing to use in this case.

All this thoughts and countless confused discussions are results of the somewhat hapless implementation of this crawler directive, and it’s ongoing semantic morphing. Sure, rel=nofollow is a generic mechanism to do a ton of things which all make sense. But it’s a geeky tool for search geeks, not a suitable tool for webmasters, editors, or publishers.

Ok, back to PageRank. Think of PR as a statistical approach to emulate Joe A. Surfer’s behaviour, where the middel initial stands for Average. PR just tries to follow Joe’s footsteps on the Web, sometimes guessing whether Joe will click link A or B, thus deducting both possible click-paths’ scores equally. PR is a model of the Web, kinda road map from any point to any point, weighting each and every path/lane’s quality by scoring the destination. And because it’s just a
model, it’s weight as ranking factor is not that important as the toolbar suggests. The PR hysteria reminds me of the Beatles BTW. Sure PR is great, it’s sexy and all that, but it doesn’t show the (whole) big picture one has to think of when it comes to optimizing contents. Hence optimize for visitors, remember the visitors pull out the plastic, not the bots, create broad and easy to walk paths through your contents all leading to your signup forms. PR will follow the visitors to honor your efforts not only with green pixels. IOW since
todays PR is Google’s secret sauce, and there are other important link cargos, it just makes not much sense to speculate about PR distribution. PR is addicting and distracting, hence you should ignore it to get your job done.

Sebastian
http://sebastians-pamphlets.com/

Admin Aaron

I agree with a lot of what you guys say but let me ask a few more questions:

JLH - If Google punishes sites that hord PR why does Wikipedia (educational), Amazon (affiliate) and other eCommerce sites doing so well in search engines?

If you look at http://solar.rain-barrel.net and any other of my sites you will see that I am doing nothing to game search engines at all but also notice the PR, how much do I have to share? >> Is PR finite?

<< Sebastian - You said” “In the case of linking to a site plastered with affiliate links and ads, where is the point to write about them in the first place, if you don’t like them?”

Say I find a farmer/inventor who doesn’t have a website but his new wind turbine was written about on a local news site that is littered with paid links and it’s “SEO” is linking out with hidden text to viagra spam. I want to write about this man who is offering something unique but the only reference I can find is on this local spammy news site. Wouldn’t this be a case when the nofollow tag should be used?

Here is another one: Say I write about a cool company that is looking to market complete solar arrays but when I review the company I find that they have multiple sites and I just do not have time to analyse their intent, wouldn’t it be safer to use a nofollow here?

I do not spam or use the nofollow to spam BUT if I understand it correctly what it does is break the connection between you and the possible bad neighborhoods that can have indirect negative impact on a websites “trust” or whatever you want to call it.

One more thought: Did you guys ever think that our debates actually build better algorithms? Are there people in white coats with clip boards taking notes? You guys got all kind of interesting ideas, that’s fo’ sho’, thanks!

Sebastian

If you link to linkworthy content you should not use rel=nofollow. If Google doesn’t think the linked site is worth indexing, that’s fine, but it’s not your problem. You’ve done your job by providing your visitors with external content you honestly think is interesting for the crowd. Trying to guess what Google could think with the intention to castrate a link when Google might dislike the target sounds just not right, and it makes no sense, and Google doesn’t encourage you to handle rel=nofollow this way. This goes for both of your examples.

Also, not every site carring tons of ads is a bad neighborhood, think of high ranked sites like SEW for example.

Would you link to my articles (I know you did it already)? Oups, I should have told you that Google (AdSense) pays my hosting fees …

ounds weird, eh?

Sebastian

JLH

Great discussion so far, probably since we are all just throwing out ideas and no one has taken a stance, so I’ll continue best I can.

I’m not sure if they do or would punish or even consider the use of nofollow in a sites ranking but for the examples, wiki, amazon, ebay, etc the hard part in that they already have a ton of momentum as far as traffic, links etc that they would be unharmed by suc a filter. I think were you’d see it is average joes site with 18 incoming links and 125 pages in the index.

Back to the finite PR discussion. I’m trying to get my head around what the question is. Are we concerned that give a pages actual (not toolbar) page rank that it only has so much PR that can go around? So if you’ve got a page with PR5 and you have 3 external links do those links receive 1/3 of your linking power, or if you’ve got 100 links do they see 1/100th of the linking power? From what I understand the actual PR of page is just related to the incoming links and has nothing to do with on-page or linking habits. A page with 100
external links and a given X incoming links will achieve the same PR of a blank page with no text and not links receiving the same quantity/quality of X incoming links.

Now the question becomes is using nofollow on a link reduce that drain on the linking power of the page. If the above is true and our page with 100 links were passing 1/100th of its linking power to each page, if we changed half of them to no follow would the remaining get 1/50th of the linking power and all 100 get the traffic flow? I’m not sure its as straight forward as that. As PR passing has as much to do with being on topic as the actual PR of the page. We’ve all seen sites that cannot even get indexed when they say that they’ve got hundreds of links, well further inspection shows that they are all junk links,
exchanged directories, off-topic etc. This has got to have the same effect on the linking page. Given the same page with 100 links and 50 of them are off topic does the page not pass PR which also means that it doesn’t drain the PR passing to the other links or does the opposite happen where it actually costs you since you choose to link to bad neighborhood or an off-topic arena?

My theory is that they currently just treat a nofollow link as an “ignore function” such the link is no longer a valid link, even though it appears to be to the general public. Google doesn’t even appear to crawl said links, on the other hand Yahoo! will crawl them for darn sure, I do not know if they consider the link in their ranking as I don’t spend much time analysing what some 3rd rate search engine does, other than send terribly targeted visitors.

My cynicisim on nofollow also comes from the evolution of it. Originally introduced to try to reduce comment spam, which it didn’t, now has evolved into protecting links for money etc. I find it strange that they got all the CMS out there to institute in for automatic use on blog comments for example, but there is no discussion on removing it. If I write a blog post and 50 people comment on that efficitivly writing my content for me, some of those comments will probably contain good links that are germane to the conversation and should not be nofollow any more as they can only help my page rank for its rightful terms. But there is no mention of that at all.

I’ve taken it upon myself to institute a nofollow policy on my own blog which I publish, its fluid and may change with time, but the key component of it right now is that I’ve used a plugin to make nofollow comment links turn to real links in 14 days. It gives me two weeks to judge each comment and if I see a link I don’t like I’ll break it, otherwise I let it mature and become a normal link that ads to the page and discussion. I’m not so sure that my policy is in line with
googles ideal, but until they give further directions its the route I want to go.

The blanket statement that all links that are paid for does trouble me a bit, based on history more than statements by such pundits like Matt Cutts. I agree in principle with what he says in that you shouldn’t be able to influence the search engine results because you went out and bought 500 links from probloggers out there, but then again they don’t have a problem with yahoos $299 review fee do they? Based on the fact that yahoo doesn’t guarantee inclusion. Well if I’m the expert on solar energy and you want me to write an article about your service that may or not may contain a link to your site, who’s to say that I didn’t receive 25 other offers from other solar energy sites
that I declinrd. In which case I think a nofollow is an misapplication of the tag. Sponsorships are a part of this society and I don’t think it can be controlled with a policy like this. Should the US government require a blanket statement after every product endorsement on TV, a warning label that says Tiger is paid by Nike to use their golf balls so please don’t accept his endorsement because of that? No we leave it up to the general public to make their own decisions. Well in the case of a paid review/posting the endorser is the writer and the general public is google. Their job is to consider the source and consider the subject and decide if the endosement carries any weight. Just as I’d rather take my golf ball advice from Tiger than Al Gore, Google should consider a paid review on Solar Energy by one of the leading experts in the field more
important than my sisters “I love Cats” blog. They are asking us to do their job judging links and pages, which just of course lends itself to even more manipulation than it was originally instituted to contend with.

Speaking of which we should really take this somewhere where someone is at least getting credit for the writing, as I’m not getting any link love or clicks on the ads to the right of this post!

Admin Aaron

Well, to be safe I do not do anything online that is “paid” other than reviewing a few products to pay my hosting fee using amazon.com iframes and google adsense that monetize a few dallors a day. I got a couple directories that scare people away with their high price and extremely strict guidelines (note, know of any good sites? submissions are free, shhh)

Great points, but one question about the Yahoo! directory, people keep using it as an example and I am not so sure it is passing link love at all. In fact, any directory that is paid would be violating Google’s guidelines correct?

Anyone know of a way or tool that shows if a “paid” directory is passing pagerank? This would be a good way to determine is Google is treating everyone equal which is another accusation I often read on blogs.

Also, how about Google just update the guidelines to show proper uses of the nofollow tag?

Just a few more ideas…

JLH

I’m not sure on the yahoo directory as I’m not paying diddly doo for anyone to review my site, but Google still references it by name in their guidelines.

“Submit your site to relevant directories such as the Open Directory Project and Yahoo!, as well as to other industry-specific expert sites.”

http://www.google.com/support/webmasters/bin/answer.py?answer=35769

Whether thats the free stuff or the paid one I don’t know. And don’t get me started on that scam of the ODP, anything that claims openness needs to have transparency and they have none. If you submit a site for inclussion and it gets shelved because the editor is your competitor that should be made public so the reasons a site wasn’t included.

I’d suppose John Softplus has some experimental data to back up yahoos linking power.

mpilatow

The official Google policy on buying links seems to be it is fine as long as you are buying links for traffic and not for PR or ranking purposes. Now how they can determine intent is debatable. Some sites make it clear that they are selling links for the purpose of PR but there are many sites who do not mention it at all but you have to know most of the people who have bought links did it to improve their SE ranking and PR score.

JLH

I stumbled upon an interesting item, I wonder if it’s related to this
subject.

http://www.google.com/support/webmasters/bin/answer.py?answer=34397&c…

Most notably:

Although Google crawls billions of pages, it’s inevitable that some sites will be missed. When our spiders miss a site, it’s frequently for one of the following reasons:

* The site isn’t well connected through multiple links to other
sites on the web.
* The site launched after Google’s most recent crawl was
completed.
* The design of the site makes it difficult for Google to
effectively crawl its content.
* The site was temporarily unavailable when we tried to crawl it
or we received an error when we tried to crawl it. You can use Google
webmaster tools to see if we received errors when trying to crawl your
site.

I’m looking at the first one in particular, where they seem to insinuate that your site may not be included in the index if doesn’t include “multiple links to other sites” Would NOFOLLOW on all external links constitute this?

Also, note to Google Editiors: The second one seems a bit dated as Google no longer runs one big crawl and a general new index push anymore as its continous. So if you want to fix that up and give me a link on your homepage that would be great.

wreilly

Is that just a typo and should be “from other sites…”

Interesting disscussion but why does it matter? PR doesn’t seem determine serp position. And no one but the bot minders know anyones actual PR.

Sebastian

Forgot to add:

You cannot lose PageRank by linking out. Google gives you PR with the duty to spread it. With every new link on a page you lower the portion of PR it sends to its link destinations, but all the PR you’ve earned from your inbounds is assigned to your pages. PR is somewhat sticky. This goes for internal links as well as external links. The only way to lower a page’s PR is to lower the PR of at least one page linking to it. Well, Google can nullify PR, but that’s done with the worst offenders only, and it can spoil down the ability of a page to bequeath PR, what you mostly don’t spot because it’s PR doesn’t change.

PR hoarding is creating a black hole, like Wikipedia did recently. Such a black hole sucks PR, but distributes PR to own pages only. BTW Wikipedia has asked Google for the permission to become a black hole before they nofollowed all external links, so they have got some special treatment with regard to PR hoarding “penalties”. Since Wikipedia links didn’t carry much weight before IMO, this black hole should affect only pages living off from Wikipedia inbounds.

Sebastian

JLH

Agreed on ranking, PR is less of an influence, but keeping your site crawled on a regular basis has everything to do with PR, which is my only concern. And if you ever try to launch a new site, having a little PR to start the crawling doesn’t hurt either.

Sebastian

OK, time to debunk some PR myths.

John, add a dampening factor and a few gimmicks to your post and you’re spot on.

{Aaarrrggg … Googlegroups has stolen my cursor again}

PR has absolutely nothing to do with relevancy, so unrelated or off topic links carry the same PR as on topic links.

{Google I want my cursor back!}

Google encouraging YAHOO paid as well as unpaid listings has not that much to do with the payment itself, but with the editorial character of the directory. So Y! links are treated exactly the same way as DMOZ links. BTW you may have spotted that some ODP categories aren’t bothered with PR, these links don’t pass PR, probably because the editors are your competitors.

{Google return my cursor or I reveal that you’ve scanned the brains of all ODP editors to figure out who’s linking honestly and who does not!}

Crawling schedules are more or less completely driven by PageRank.

High PR values result in frequent crawls, low PR makes Ms. Googlebot lazy. Laying out milk and cookies attracts and holds Ms. Googlebot, she loves cookies.There are other factors like freshness and source scores involved when it comes to fetching brand new stuff.

{Nasty Google, I do miss my cursor … pleeeaaase! }

“Page” in PageRank stands for Larry Page, not Web page.

The PR of the whole Web is 1, not 42.

Sebastian out to hunt for a new cursor

Admin Aaron

LOL, thanks for all the info. today but I must say, none of us still know for sure if PageRank is finite and can be used up like fuel.

Sebastian >

Definitively finite, PR cannot leave the Web.

JLH

The law of conservation of then PR applies? Page Rank can neither be created nor destroyed. So PR is more like energy which is just converted and transfered than like currency which can be produced, spent, wasted?

softplus

So who’s going to mirror this discussion in their blog so that we can link to it?

Great stuff, guys.

Is pagerank diluted as you add more content to your website? Is pagerank diluted as Google adds more filetypes to the index? Can issues with numerical accuracy result in “chaotic” pagerank fluctuations?

John

If you frequent webmaster forums, discussion groups, or blogs, you’ll more than likely see post after post, and question after question regarding PageRank (PR). For those who don’t know PageRank is the little green bar up in your google tool bar that “is Google’s measure of the importance of this page.” The scale is a simple 11 number 0 through 10. Being that its the only visible indicator from Google it’s often obsessed about.

For a discussion about PageRank, we first need to look at the definition by Google.

PageRank relies on the uniquely democratic nature of the web by using its vast link structure as an indicator of an individual page’s value. In essence, Google interprets a link from page A to page B as a vote, by page A, for page B. But, Google looks at more than the sheer volume of votes, or links a page receives; it also analyzes the page that casts the vote. Votes cast by pages that are themselves “important” weigh more heavily and help to make other pages “important.”

Important, high-quality sites receive a higher PageRank, which Google remembers each time it conducts a search. Of course, important pages mean nothing to you if they don’t match your query. So, Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your search. Google goes far beyond the number of times a term appears on a page and examines all aspects of the page’s content (and the content of the pages linking to it) to determine if it’s a good match for your query.

Top 13 things that won’t effect your PageRank

Though that seems to sum it up pretty good, often I still see numerous questions regarding a sites page rank going up or down. To start with let’s cover some common elements that have ABSOLUTELY NOTHING to do with a pages PR calculation.

1. Content
William Shakespeare himself could come back to life and write for your site and Google’s page ranking won’t give you one point for the quality of your content. You can spend 20 hours a day for 5 years writing thousands and thousands of unique, highly relevant, beautiful pages and your PageRank will not change. Conversely you can have blank pages and your PageRank will not change.
2. Titles
Changing your page titles from all just the name of the site to a unique description of each page will not improve your PageRank. Erasing all of your titles will not decrease your PageRank
3. Description
Including a highly detailed and extremely accurate description of your page will not affect your PageRank, if on the other hand you accidentally remove all your descriptions and replace them with the number “7″, your PageRank will not suffer.
4. Keywords
Whether you use META keywords on your page or not, or your keywords are stuffed, rotten, or totally irrelevant has nothing to do with PageRank.
5. Google Adwords, Adsense, Analytics
The use of any of Google’s other services including AdWords, AdSensse, or Analytics has nothing to do with PageRank. Whether you spend million dollars a day with AdWords, receive a million dollars a day from Adsense, or check your stats every 10 minutes, has absolutely nothing to do with your PageRank.
6. Sitemaps
Using Google’s webmaster tools and submitting a sitemap or unsubmitting one will not effect your PageRank.
7. Sitemap Errors
If Google’s webmaster tools show you have no errors or 500 errors in your sitemap, your PageRank will not be hurt or helped.
8. Server Problems
If your server goes down for a week or has never been down since Al Gore invented the internet , your PageRank will remain the same.
9. Competitors
Your competitors use of sneaky redirects, hidden text, adwords, sitemaps, the color green, linking scheme, SEO budget, cloaking, affiliate links, or page rank will have no effect on your PageRank.
10. Robots.txt
Having or not having a robots.txt that works or doesn’t work will not help or hurt your PageRank. You can block Google from your site entirely and you will still get PageRank.
11. Duplicate Content
Whether you write all of your own content, have a scraper write it, someone else copies it, publish the same page 500 times, your PageRank will not be harmed or icreased.
12. Age of Domain
If you just bought your domain or got it from Al Gore back in the seventies, it has no bearing on your PageRank.
13. Images/Flash/Java/CSS
You may have the ugliest site in the world or the most beautiful flash laden java enabled page with pictures taken by a professional photographer and it will have no effect on your PageRank.

The point is that only the links and the quality of the links is considered in your PageRank calculation, that’s it, nothing else. All of the above items may help get you links which will increase your PageRank, but they will not be considered in the calculation. You may find an old crappy site that doesn’t have any links but has high PageRank, but you have to consider that that link may have been there for 10 years.

PageRank Factors

Now that we’ve got what doesn’t count out of the way, let’s consider the remaining factors that do.

1. Quantity of links
Of course any calculation must consider the shear quantity of links. More links is better than less links. Less links is worse than more links.
2. Quality of links
Links from higher ranking pages are given more weight than lower ranking pages.

Unknowns

Most often when one asks, “what happened to my page rank?” or “why don’t I have any PageRank?” it is because of these unknowns.

1. What is a link?
We don’t know exactly what Google considers a good link. Some sites that link may have had their link power taken away. Other links like NOFOLLOW links don’t carry any weight. Even in the webmaster tools the links shown may not carry any weight. Using Yahoo’s site explorer to find links to your site has nothing to do with what Google considers.
2. The calculation
If we definitively knew what links Google considered we still don’t know how they calculate how much weight each one passes on. It has to have something to do with the page rank of the linking page, the quantity of links on that page, and other factors but the exact calculation is unknown.
3. Internal PageRank
The page rank you see in the tool bar is not updated often, only a few times a year and even then pretty poorly. If you check multiple data centers you will see that even Google isn’t on the same page as to what a page PageRank is. The internal PageRank however is a continuous function that’s updated while they crawl and index the web.
4. The Scale
There are a few billion more pages today than there were in 1998 when this whole page rank mess yet it is still ranked 0 through 10. Back in 1998 maybe you only needed 100 links to get to a page rank of 6, but today you may need 10,000 links. As the size of the web grows the threshold for the next level in PageRank changes.

Conclusion

PageRank is simply a calculation based on the quantity and quality of links that Google considers to a page. The exact calculation is unknown (to us outsiders) and is ever changing. Most important however to know is that a page with page rank of 2 can and will outrank a page with a PageRank of 6 in a search results page. Given that simple fact, if you’ve read to the bottom of this page then you’ve spent too much time already thinking about PageRank, spend more time with the basics like making sure it’s crawlable.

Update2/14/07

In response to Monika M’s comment below I thought I’d show an example of a lower PageRank page, outranking a higher page. There’s a great little Firefox extension for SEO by Aaron Wall that will let you see the PageRank and other important data about a page right in the search results. Using this extension turned on I ran a search for “webmaster” on google. You will notice that the #1 spot is a page rank of 6, outranking the #2 which is google itself at #9.

Regarding the second part of your comment, I think non-reciprocated links are probably the BEST link you can get pointing to you, it helps define the site as an authority. Think of the great page ranked sites out there with millions of links TO them and very few that are reciprocated (Ebay, CNN, Amazon, etc) . I wouldn’t lump all reciprocated links as bad however, if the two sites relate to each other than it’s a good thing. If you’re selling jewelry and De Beers links to you on the their home page, I wouldn’t be afraid to mention it on your site. Who links to you, as well as who you link to helps establish your keywords that your site will rank for, along with on-page content. It’s this simple fact that was taken advantage of when people started Google Bombing. If a site is on a related theme that you are, linking to them will help your site. This is why reciprocal linking just for the sake of getting a link is usually bad if off topic. Unless you really want to rank for “Free Ringtones” or “Cheap Mortgages” having them link to you and you linking to them is only bound to take your sites theme off course. That being said, don’t forget that there is virtually nothing anyone can do external to your site to harm your rankings. So if you find a bunch of links from some off topic site to yours, it’s not going to hurt you, as soon as you link to them however you’ve joined their “neighborhood” which may not be so much of a good thing.