The Truth Behind The Big PageRank Slap.
Recently Google updated the PageRank ratings for websites and a lot of sites saw their PageRank plummet. Many webmasters have speculated that Google “slapped” sites who were selling links for the purposes of passing PageRank… but does that theory stand up?
Let’s look at it from Google’s point of view…
If I were Google and I didn’t want sites selling links that passed PageRank, I would do three things…
1: Encourage any sites selling links to use the “nofollow” attribute as part of the link. Making this suggestion places the emphasis on the *site owner* to comply with Google, which naturally saves Google a lot of time and effort. Perhaps “model” sites like this would get a small ranking boost…
2: Modify my algorithm to identify and downgrade any links that appear to be links purchased for PageRank. Sites who display “Sponsored links”, “Advertiser content”, “Ads” etc can be used to identify patterns in ad location and content. Any patterns identified can be fed into the algorithm and used to locate sites who are selling ads but *not* disclosing the fact to their visitors. If those links don’t contain the “nofollow” tag, then the site is selling links that pass PageRank.
I would then ignore those links for PageRank calculations… especially if they are to sites whose content isn’t related to the content of the page where the link is found.
At his blog, Matt Cutts said Google could detect paid links…
“Google has a variety of algorithmic methods of detecting such links, and they work pretty well.”
… then he gave an example …
“when dailycal.org is selling links to casinos or other link buyers, we prefer not to trust those links.”
… and finally Matt refers to “an email from a “text link broker” that included this suggestion:
Most people use words like, SPONSORS, PARTNERS, FEATURED, ADVERTISERS, ADS and other synonymous terms related to advertisers. Our suggestion is to use ‘different’ titles for these ads. Something like RELATED SITES, COOL SITES, RESOURCES, ALTERNATIVE LINKS and so on.
The email later suggests “to use unique locations for ad links like within content.”
To which Matt says… “I wouldn’t be surprised if search engines begin to take stronger action against link buying in the near future.” Matt Cutts Blog, September 2005.
Not surprisingly, that comment now seems quite prophetic.
3: Downgrade the whole PageRank system. If selling anything up to a PageRank 3 or 4 link was not worth the effort, downgrading all sites by a few PageRank notches would make the business of selling Pagerank disappear almost overnight. All Google would have to to is increase the stratification of the PageRank system… so that more sites are in the lower ranks, and the selling of PageRank would decline substantially. Of course, downgrading Pagerank across the board would just mean that a PR5 page may be equivalent to an “old” PR7 page… but who’s going to want to pay the same amount of money for a “new” PR5 link compared to an “old” PR7 link? And if they buyers disappear, the business of selling PageRank links dies too.
A litmus test for this theory is whether or not your Google traffic falls with your PageRank. My tests show that it doesn’t… which would lead me to believe that the whole system was downgraded. If the loss of PageRank doesn’t affect the amount of traffic coming from Google, then the PageRank change was cosmetic.
I believe Google has done all three… they’ve told sites selling links to use “nofollow”, they’ve most likely identified advertising blocks (especially to “non-relevant” sites) and downgraded those links. They may well have downgraded the whole PageRank system to make selling PageRank links less attractive. I also believe they hit “article directory” sites and downgraded their outbound links too.
One blog comment post I found said, “Actually, if you look at it, it doesn’t seem like any sites are above 5. Maybe 5 is the new maximum.”
I think that guy spoke the truth more closely than he realized.

November 7th, 2007 at 2:07 am
Google certainly keeps us on our toes. Looks like MSN, Yahoo, and Google are still an 8, 9, and 10. I noticed the fall of a few sites that were 5 or 6.
Nice suggestion on the alternative labeling of our links.
Should be interesting to see how quickly the rise in PR will be for the “demoted” sites.
Ted
November 7th, 2007 at 6:11 am
I noticed that most of my own sites have dropped by 2 - and yet I came across a few sites yesterday while researching some stuff that had not changed in PR and I’ve always wondered how they got the PR they had in the first place. I find the whole PR thing quite puzzling and simply ignore it.
I don’t understand how calling a link something else changes it function? Might as well call smack recreational hallucinations.
Kevin
November 7th, 2007 at 6:21 am
My site on how I’m losing weight actually went up. I was 0/10, and now I’m 1/10. This happened in the last couple of days. Alexa has finally picked me up too. So mine was not downgraded, that I know of. But I can understand why they did.
November 7th, 2007 at 10:36 am
The other thing I find a mystery is why google, yahoo and msn do not pick on all links from articles posted around the net. Yahoo and msn “credit” my site with more article links than google, so why does google discount so many of my articles?
I post a unique article to about 15 directories so it is not mass blasting of the same article. Some articles of mine link back to me from sites that are not article directories, and even those links are not picked up.
Any advice?
Thanks
Shelley
November 7th, 2007 at 10:58 am
Hi Shelly,
For a long time, Google haven’t been showing *all* the backlinks they know about when you query their search engine. They only show a selection of backlinks. That’s why webmasters query Yahoo to get a good picture of a site’s backlinks.
But Google will know about all your backlinks… even though they don’t display them all.
Neil.
November 7th, 2007 at 12:05 pm
I havent had any problems with this. So I wonder if google is “targeting” specific types of blogs and paid links. I got an upgrade in page rank actually.I’ll be happy when the algo’s start mimicking what happens in real life though. If ranking on google is dependant on back linking then wouldn’t it be nice if this voting “word of mouth” actually was dependant on links in posts and content pages and not just side bar activity or paid links. Either way with every slap there is a work around by some crafty SEO geeks. Seriously though, imagine when you buy a magazine if the publisher put all the ads in it at the back of the magazine. What would happen to the mag revenue and the economy in general. Or the same with TV spots. (ok I like commercials) But what would happen is the networks penalized the stations for having ads in their air time blocks.
Google is just stupid sometimes, time for them to get out of their ivory tower and start getting into the streets and seeing what’s happening in the real world
November 7th, 2007 at 1:34 pm
I happened to be upgraded from PR 2 right to PR 4 just a few days ago. I am wondering whether besides collecting good back links - which surely helped - also having a few PR 5 and 6 domains that I could buy cheaply (not related to my subject) point/forward to my hopepage did help. Does anybody made experiences with such “High-PR-domain-Forwarding”?
November 7th, 2007 at 1:49 pm
Hi Juergen,
I’ve thought about buying sites as PR boosters for other sites… I guess that’s taking buying text links one step further… buying the whole site.
Neil.
November 7th, 2007 at 3:54 pm
I really don’t put much weight in pagerank myself. To me, it’s more an indicator of how frequently the Googlebot will visit my sites and an indication that my new pages will be indexed quickly. But many folks value this as an indicator of their website’s success. A new client of mine quickly put a check in the mail to me after his site went from a pagerank 0 to a pagerank 3 after this last Google shift. To me, traffic stats, organic linking, and conversions are more of an indicator of a website’s success.
November 7th, 2007 at 4:38 pm
Hi Jim,
I agree completely. It’s like the marketer who puts a sports car on their sales letter to indicate “success”. But, the “show” is important too… people’s first impressions of you are important… whether that means a flashy car or a PR9 homepage… it’s still affects how people perceive you and how “successful” you are.
Hmmm… I wonder if a flashy car can be a business expense? Actually, it’s my business and it won’t be buying any fast-depreciating flashy cars any time soon!
Neil.
November 7th, 2007 at 5:32 pm
PR is now as valuable as Alexa rank!
November 7th, 2007 at 5:57 pm
I may be misguided here but surely as more websites are added to the rankings, surely it’s harder to get a high pagerank. I would assume that the pagerank calculation will be based on a critical mass or websites each time it is recalculated, as a control referrence of who shoul get what PR compared to someone else. So basically if you are not building PR on a monthly basis your PR is going to be going down. I assume that at some point only the biggest sites will have any PR at all. Would that be correct Neil?
Warmest Regards
Mark
November 7th, 2007 at 6:17 pm
Hi Mark,
I think Google index every single page they can, then use the total number of pages as the PageRank “pool”, then apply their algorithm to decide what PageRank score individual pages should have. In that manner they align their system with an increasingly large web index.
I seem to remember Leslie Rhode saying that each page you create generates a tiny amount of new PageRank… which basically means the same.
So, if the web is growing, but your site isn’t, your “share” of it will fall proportionately.
Neil.
November 7th, 2007 at 7:50 pm
Neil, that’s an interesting way of looking at it.
So if the total PageRank of all web pages is indeed a finite number, that makes competition for PageRank much more fierce.
Hypothetically, if the grand total of all PR across the internet was 100, and there were only 20 or 30 pages in existence, there would be plenty of PR to go around. But if there were suddenly 10,000 web pages there would be very few with high PR and the rest would be left to clean up the scraps.
Tthat’s just an example using tiny numbers compared to the billions of web pages that exist, but with new sites going up constantly there must be a breaking point at some point or another.
Of course, I put little stock in PR anyway. It’s better to concentrate on creating quality content that brings in revenue than worry about a little green bar in your browser. Though you make a good point about a high PR making a good first impression on visitors. Something to think about.
November 7th, 2007 at 9:38 pm
Hi Mike,
I think it’s more like the “Total PageRank Pool” is equal to the total number of webpages indexed. So, when the web was 100 pages, the sites shared a pool of 100… but now the pool may be 10 billion. Some sites with high PageRank will gain links as the site grows and maintain their high ranking… other will lose out to new sites who are growing faster… but it’s growth relative to the overall growth of the pool that’s important.
For more on PageRank, see…
http://en.wikipedia.org/wiki/PageRank
Neil.
November 8th, 2007 at 11:00 am
Well for me I think the whole PR thing is more of Google way of ’showing the big image’ because I really don’t see the benefits of higher PR above a lower PR in terms of traffic and sales.
I have a site that was PR0 and now PR2 but traffic and sales remain the same via Google. What then is the whole fuss about high PR all about? About selling links, I also do not subscribe wholeheartedly to it. I prefer building links in natural formats.
Thanks.
November 8th, 2007 at 12:32 pm
The PageRank toolbar update is generally considered to not reflect an instant change in your site’s PageRank, but the PageRank of what your pages were weeks or possibly months ago.
There are important other reasons for high rankings such as the link reputation and the keywords on your page… but the PageRank of the page also plays a role in determining where your pages rank in the search engines for particular search phrases.
As the PageRank scale is logarithmic, there is very little difference between PR0 and PR2.
Type “business” into Google and the first 5 results are PR 7,8,6,7,8. The first is, unsurprisingly, business.com.
Neil.