Is Google Really Planning To Put PageRank Out To Pasture?
October 4th, 2007
You’ve probably heard the rumors — you’d almost have to be deaf to not have — that Google may be ditching Page Rank, or at least the public display of PageRank on the Google Toolbar.
In the interest of full disclosure, we freely admit that we have no more idea than anyone else whether the rumor is true. And, frankly, we tend to doubt it since PageRank is one of Google’s very best, most sublimely useful tools for creating FUD (Fear, Uncertainty & Doubt) among webmasters.
If the rumor is true, however, the demise of PageRank would rank as a textbook example of “good riddance to bad rubbish.” Rubbish? Rubbish!?! How dare anyone call Page Rank rubbish?
The better question is how dare anyone — with “anyone” defined as a Google user — not call it rubbish? Other than, of course, the usual mob of people who always see the emperor tricked out in his ermine robe and diamond-encrusted crown even when he’s jaybird naked as the day he was born.
Consider automobiles.
If you Google “automobiles” you get over 56.5 million returns … all of them bearing a PageRank of 0 to 10. Doing the math, you discover that if there were a bell curve dividing the 56.5 million equally among the PageRank numbers, there would 5.65 million sites ranked zero, 5.65 million ranked ten and 5.65 million ranked every number in between.
There is no bell curve, of course, so what you probably have are only a million or so zeros and tens and, maybe, 10,000,000 sixes. Discount, for a moment, those who say a return position past the first two pages is useless. For the sake of argument, let’s say even page 20 is great — which is true, in an abstract sense, when you’re dealing with tens of millions of returns.
Even better, call page 100 a great return position. With millions of sites having exactly the same PageRank vying for the first 1000 returns, what possible meaning can PageRank have in the real world of getting hits and making sales?
The first return for “automobiles” on 10/1/07, a New York Times review, had no PageRank. The second return had a PageRank of 5, the third a PR of 8, the fourth a PR of 5, the fifth (automobilemag.com) a PR of zero, the sixth a PR of 6, the seventh (an FTC page on cars) a PR of zero, the eighth a PR of 7, the ninth a PR of 7, and the tenth a PR of 4.
OK, maybe — not likely, but maybe — there are no pages related to automobiles that have earned PageRanks of nine or ten. Even so, there still must be millions of pages with ranks of eight, seven and six. Yet in the place that matters –the search return roster — you find three zeros, one four and two fives in the top ten. Do that math and you discover that in a search for “automobiles” over half — 60 percent — of the first ten sites returned have Page Ranks of five or under.
Perhaps “automobiles” is a bad search term for this kind of test. With over 50 million returns it may be atypical. Let’s try something more manageable, like “Chicago Real Estate Brokers.” That’s better, only 1503 returns.
Here are the Page Ranks of the first ten returns in order: three, three, four, one, two, two, one, two, zero, three.
Oh, dear, not even one site with a PageRank as high as five. Perhaps these webmasters should all fire their Searchh Engine Optimization consultants and bring in a team who can get them higher PageRanks and return positions a few hundred spots back from where they are now.
Or, perhaps, their competitors — the ones with high PageRanks and low return positions — should just spend a bit of time and effort enhancing and expanding their link development campaign, which can be of crucial importance in the key aspect of search-engine optimization: Improving your return position.
The reason for this is simple: The quality and relevancy of your links and the information included in the links’ anchor text are among the major factors Googlebot considers in deciding which of the hundreds, thousands, or millions of sites with exactly the same PageRank wind up on return page one and which are dumped on return page 1000.
So what do you think? Should Google dump PageRank? Should it retain it as is? Or should it replace it with a system that gives each site its “real” rank — from first to 56 millionth in the case of “automobiles.”
Post your comments below or in our Discussion Forums and let us know.


November 13th, 2007 at 12:37 am
AFTER PAINSTAKINGLY OBTAINING LINKS AND BUILDING UP MY LINKS PAGE TO AT LEASE A PR 2, I FOULD THAT IT IS NOW A BIG, FAT ZERO. PAGE RANK ZERO WILL DEFINITELY AFFECT THE QUALITY OF LINK PARTNERS THAT WILL REQUEST TO LINK WITH MY SITE. THE ONES WITH HIGH PR WILL REFUSE TO LINK WITH OUR SITE. DO NOT KNOW HOW TO GET OUT OF THIS ONE AND WILL BE GLAD IF MR. CUTTS ANNOUNCE THAT IT IS “GOOD RIDDANCE TO PAGE RANK” FOR GOOD.
December 4th, 2008 at 2:52 pm
“If you Google “automobiles” you get over 56.5 million returns … all of them bearing a PageRank of 0 to 10. Doing the math, you discover that if there were a bell curve dividing the 56.5 million equally among the PageRank numbers, there would 5.65 million sites ranked zero, 5.65 million ranked ten and 5.65 million ranked every number in between.”
No, dummy, that would be a flat line. Not a bell curve.