Forum Moderators: open
It's my first time in a webmaster's forum since I am not a webmaster :). I did launch a business content-rich web site 14 months ago and although I am not a web developer (I am the managing director but have some basic HTML knowledge though) I started looking for search engine optimization techniques six months ago.
I have read many articles and papers and yesterday I read the outstanding paper "PageRank Explained or Everything you?ve always wanted to know about PageRank" <edit>
that had some very interesting points to show me.
I would like to share with you our case study and request your valuable feedback.
Here are the facts:
Our web site (lets say it www.site.com) holds a PR6 and has currently 980 backward links in Google (about 70% from our site's internal pages). Now it holds the position #21 for a single search term that shows 2,800,000 results in Google - I hope that's not bad for a non-webmaster! :).
Our target is to move our site to PR7 or PR8 until the end of this year. Currently we are at the end of a reconstruction project. In the next few days, we will have about 4,200 static pages (news, papers etc.) while we used to have them mostly as dynamic URLs (with ?).
We also operate and handle 7 more sites (5 in .com and 2 in another country) that have an average PR5 but do have less than 50 static pages each. These 7 sites belong to companies related to our site's content (critical issue, don't you think?).
We would like to exploit these 7 domains in order to enhance the PR of our site by doing the following (??):
1. More indexed pages: We will use the Extensive Interlinking structure (as shown on the paper I mentioned above) for all the main sections of our site (approx. 20). In fact we will provide all our pages with a footer table with text keyword-rich links to all our main sections.
2. Not deep folders: We will not use deep URLs and certainly not more than 2 roots deep.
3. External links: We maintain a directory where companies from our industry may post their link (we do have many of them). We will use the Hierarchical Linking model for these pages in order not to decrease the PR of our site. Additionally we will use javascript links (I think they are invisible) for all the rest external links on other pages of our site (they are not many).
4. Exploiting the other 7 sites: We plan to publish all our news pages (around 3,500) as static pages on all 7 sites and publish links to all 7 home URLs on all of their pages. I hope that this will increase the PR of the 7 sites. We plan also to link all their pages to the home page of our site (www.site.com). Our site will NOT provide links to them in order not to decrease its PR.
5. The meta tags of the 7 sites (and all their pages) have been modified in order to match better our site content.
6. The text links that we will use for cross linking between the 7 sites will have the main keyword for our site (is also a keyword for the 7 sites as well).
The main questions that occur are (I know there are many more!):
1. Is the cross linking method a good way to increase the PR of the 7 sites?
2. Could the content exchange be considered as spamming by Google, although each of the 7 sites do have related content?
3. Is the Extensive Interlinking method the best way to increase our PR?
4. Some of the 7 sites do have some non-English content (i.e. on the home page). Does this matter for their PR?
This is our case! Since I am not an expert, I would appreciate to have your valuable feedback on such a structure. Let me finally mention again that all 7 sites do have content related to our site. So we are not trying to spam Google. I think that content exchange - I mean the publication of our 3,500 news stories on each of the 7 sites - is not considered spamming (or am I wrong???).
Look forward to hearing from you all... I would be pleased to share with you the results of our plans.
Aris
P.S. I should apologize for my English! It's not perfect but I hope you do understand our case.
[edited by: NFFC at 10:54 am (utc) on Sep. 8, 2002]
[edit reason] Url removed [/edit]
I think the only exception to your argument muesli is that if:
Main page has say PR7 -- 10 million page of your complete site were actually indexed in Google each having a link back to the main page but none have there own developed PR (inbound links), their PR is only from a transfer.Even if all 9,999,999 had a direct link from the main page so that the most PR could be transfer, I don't believe you can increase to a higher PR level (PR8) because the external influences are not there to support.
the PR scale is logarithmic (the step from 6 to 7 is a lot smaller than from 7 to 8), so it would take quite a lot to move my frontpage (actually being PR7) up to 8. anyway your scenario would bring me some kind of PR boost. only that it is practically impossible as those 10 million pages wouldn't make it into the PR calculation due to lack of PR. even if i all linked them from my frontpage they wouldn't as google stops indexing a page at a certain file size and 10 million links would certainly create file size ;)
basing crawling depth on pagerank wasn't BIG G worst idea actually, it is a good filter that keeps people spamming the PR algo by simple number of pages (as aris intends to do).
In this noone could even get to PR10
As long as you keep getting new PR (at whatever level) from external influences you will increase.
But I agree, without any new PR coming in, your not going higher no matter however many pages, are indexed and receiving only internal transfers.
[edited by: fathom at 2:57 pm (utc) on Sep. 8, 2002]
I fully agree with what you state. When I post this message for PR, I did NOT think that this is the 1st issue in our web strategy! You are absolutely right on this. We are certainly doing more things for the other important issues you mention. But PR is the case here in this forum.
I guess you do not agree on the manipulation of the 7 sites as I suggested (I am not an expert...). Is the content exchange (or "duplicate" as many say) a problem if we decide to provide our content to the 7 sites without any interlinking? This could be a first step to see if the PR is influenced by the quantity of content available. Or not?
No, no one could get to a PR11.
Have a good look around at the 10's, almost a closed shop, look at Apple > Adobe > Macromedia etc.
Apple > Adobe > Macromedia etc.
How did they get to 10? By linking to each other?
The PR of a page can never be higher than the highest PR page that links to it. If you want to be a PR8 then you need at least a PR8 linking to you.
this is definitely not true. i've had PR7 for a about a year now and only received my first incoming PR7-link (beside links from my site's other pages) this month.
thousands of PR1-5 links can be as valuable as one PR7 link. (unfortunately i don't know exact relations though.) in my case the majority of those thousands even come from my own site - from user pages that are themselves filled up with PR by deep links from outside.
if it was as simple as you state PR simply wouldn't work. my opinion. check the paper aris mentions, it's probably not exactly how PR works (anymore) but i'm confident that the basic framework has stayed the same.
another reason: if it was as easy as you say only SEOptimized pages would get high PR. but in fact i had PR7 long before i knew what SEO stood for, even without knowing i had PR7. many low profile links certainly make a page much more an authority than a single PR8 link that might even be paid.
Link volume has always been considered a factor in Google...alright, not a factor in Pagerank math itself perhaps but a factor nonetheless.
On the other hand, PR8...well...PR8 :) Few and far between.
I'm not so sure on pages scoring higher than the pages link to it though......
Regarding duplicate content and/or cross-linking ('interlinking') - I'm certainly no expert either, but I gather that cross-linking is the greater offence in the eyes of Google. I can't really comment on the idea of duplicating your content across seven (unlinked) sites, except that it seems like a lot of work just to test a theory. I'd wonder also about your directory listings for those other sites - in the long run don't you run the risk of a de-listing from the important directories (DMOZ for example) due to duplicate content? Loss of a DMOZ listing, and consequently a Google Directory listing, could see your PR actually fall.
the PR scale is logarithmic (the step from 6 to 7 is a lot smaller than from 7 to 8), so it would take quite a lot to move my frontpage (actually being PR7) up to 8. anyway your scenario would bring me some kind of PR boost. only that it is practically impossible as those 10 million pages wouldn't make it into the PR calculation due to lack of PR. even if i all linked them from my frontpage they wouldn't as google stops indexing a page at a certain file size and 10 million links would certainly create file size
Ya, ok my example was a little off the wall but even at PR5 and internally passed PR, I really can't see going to PR6 without ever getting another external link (an external PR1 link might do it but...).
PR is from external links - so a new site, newly indexed with 1 incoming link gets a certain value and that page can just pass that value around to create a higher value... it doesn't make sense to me.
For of it's complexities math is very simple... in audio (decibels) which is also logarithmic 2, 100 DB sources doesn't creat 200 DBs but 103 DBs but adding another speaker as part of the 100 DB source(representing internally PageRank) doesn't increase the level one ioda. My understanding, PageRank is by page and site rank doesn't realy exist so the highest page can't really get any higher than where it's at without something new (external being added to the mix) all other things being equal of course
All this is nice and dandy - before I go crazy let me point out a couple of things...
The way PR works is pretty well known.
I would prefer 1 link from a PR4 page with no other external links than 5 links from PR8 that host 20 external links.
I am with NFFC - using my fancy estimations. The latter would be about 200 (and I am being conservative) times more valuable.
YOU CANNOT [for all intents and purposes] CREATE PAGERANK - STOP TRYING.
Your contribution to pagerank is the number of:
(pages google decides to index of yours)/
(pages in the google index)
If these internal linking perpetual motion machines workd - don't you think that either:
1) Someone else would have done it?
or
2) Someone else would have gotten it to work by accident.
You have power over who you distribute RageRank to. The entire rest of the web has power over who distributes PageRank to you.
Unless you can get PR to some of your pages - you are not going to get a huge amount of PR. If you get X amount to your main page - you have X amount to play with. Google is not going to index millions of subpages off your main page.
As far as the general idea that you can't have more PR than the highest page that is linking to you - it is generally true, but not because of PR, but because of statistics and human behavior:
Sites that are below PR4 you can't accurately do backlinks on.
Sites that are PR7 and above are usually legitimate sites.
Most PR7s have hundreds of links to them - this site for example has 5,400 links according to google.
With that many links - odds are - the way it works out is that there will be at least one site higher than yours pointing to it.
To think of this another way - we all know or should know - that a PR4 site is worth something times a PR3 site - probably around 6. In any event - using this further down the line - a PR3=PR2X and PR2=PR1X.
Ok - and then the links on the page have to be taken into account - now lets look at an EXTREME example:
Would it be POSSIBLE for a PR5 site to be made up of just links from PR1s - yes in theory it could?
However, for this to happen - you would have to believe that thousands of PR1 pages would link to it without hundreds of PR2s or tens of PR3s....
It just doesn't happen in the real world - and it wouldn't suprise me if this was part of the google anti spam checking machinery. Discounting for people trying to get their competitors in trouble - I'd lay my money on a PR4 site with no PR4 or 5 links to it being spam than invisible text, repeated text, and hidden links any day.
It simply doesn't work out that way in theory very often.
Just my 2 cents.
The idea is, that google will see you as a HUB of information for your particular industry if you have links. Google will then give you more importance, see your site as an authority, and find you more relevant for your keywords/industry if your site links out to other sites.
Of course, don't link out to competitors!
Attention PR Theorists:
Thanks Chris_R, a few hours ago I was just another spammer.
Does Google use the "hub and authority" approach? I know that some search engines do, but I wasn't aware that Google did. That approach may have validity (and I like it, since I have several thousand outbound links), but I don't know how it would fit in with Google's PageRank concept.
If you follow this board, you would notice the huge amounts of PR0 threads. These are not by accident, nor are they caused by silly mistakes. They are caused by many of the techniques you describe.
Of course, don't link out to competitors!"
Maybe thats why I am doing well in my niche? However, I DO link to my competition (in fact to all I know of) and it doesn't seem to hurt me (#2 on main keywords). Personally, I do believe that Google will see you even more as an authority if you have the guts to link to your competition (this feeling is based on the academic papers available from Sergey Brin and Lawrence Page when they where still students).
Honestly I believe PageRank is not the happy-maker some seem to believe, but I am fast to admit, I simply don't have a clue why Google lists me above most of my competitors. On the other hand, the majority of my traffic is still coming from web sites that link to me.
And, even though Google sends more traffic to me than any other SE, I did have to notice, that Google visitors usually are leaving my site as fast as they come - they find what they are looking for and off they go!
For me, Google brings traffic, but business comes from other sources.
BTW: I do remember 1 1/2 years ago, when I had a lot of Google visitors looking for an international airport (at that time Google didn't have Airport web sites indexed). Today not a single Google visitor looking for airports is passed through my site since Google has the Airports indexed themselves.
"Another intuitive justification is that a page can have a high PageRank if there are many pages that point to it, or if there are some pages that point to it and have a high PageRank. Intuitively, pages that are well cited from many places around the web are worth looking at. Also, pages that have perhaps only one citation from something like the Yahoo! homepage are also generally worth looking at. If a page was not high quality, or was a broken link, it is quite likely that Yahoo's homepage would not link to it. PageRank handles both these cases and everything in between by recursively propagating weights through the link structure of the web."
From "The Anatomy of a Large-Scale Hypertextual Web Search Engine" Sergey Brin and Lawrence Page, Computer Science Department, Stanford University, Stanford, CA 94305 (at that time -1997- Google had 24 million pages indexed)
[http://www-db.stanford.edu/~backrub/google.html]
Google is, at least partially, depending on human maintained lists (Yahoo, DMOZ etc.) they trust.
This means there is a human factor in determing the ranking of a site in SERPs. (wonder why microsoft has a PR of 10?)
Anybody who is trying to artificially boost his ranking by fooling the Google algorythm and their human editors (mainly Yahoo and DMOZ) should keep this in mind!
Google and his inventors have lost innocence on the road to become the most powerful SE around.
"Currently, the predominant business model for commercial search engines is advertising. The goals of the advertising business model do not always correspond to providing quality search to users. For example, in our prototype search engine one of the top results for cellular phone is "The Effect of Cellular Phone Use Upon Driver Attention", a study which explains in great detail the distractions and risk associated with conversing on a cell phone while driving. This search result came up first because of its high importance as judged by the PageRank algorithm, an approximation of citation importance on the web. It is clear that a search engine which was taking money for showing cellular phone ads would have difficulty justifying the page that our system returned to its paying advertisers. For this type of reason and historical experience with other media, we expect that advertising funded search engines will be inherently biased towards the advertisers and away from the needs of the consumers."
From: "The Anatomy of a Large-Scale Hypertextual Web Search Engine" as above!
Thinking of "Sponsered Listing" and "AdWords" one may be allowed to think about the overruling effect of power on believes!
Again: From "The Anatomy of a Large-Scale Hypertextual Web Search Engine" by Sergey Brin and Lawrence Page
"For example, we noticed a major search engine would not return a large airline's homepage when the airline's name was given as a query. It so happened that the airline had placed an expensive ad, linked to the query that was its name. A better search engine would not have required this ad, and possibly resulted in the loss of the revenue from the airline to the search engine. In general, it could be argued from the consumer point of view that the better the search engine is, the fewer advertisements will be needed for the consumer to find what they want. This of course erodes the advertising supported business model of the existing search engines. However, there will always be money from advertisers who want a customer to switch products, or have something that is genuinely new. But we believe the issue of advertising causes enough mixed incentives that it is crucial to have a competitive search engine that is transparent and in the academic realm."
Question: Why does Lloyds have to make a $1,55 Million dollar advertising deal with Google to be found by Google users?
[http://www.forbes.com/technology/newswire/2002/08/01/rtr683315.html]
The day Google started to sell advertisement on SERPs, Google turned into "just another SE". And, without need, considering the money they make selling their search solutions to enterprises for $28.000,00 - $450.000,00 [http://www.searchtools.com/tools/google-app.html].
My conclusion? Wake Up! The Internet is commercial! 100%!
Don't rely on traffic because you have a high PR or ranking at the moment. Google is a business, out there to make money and as soon as there is a competitor who is willing to buy you out - you are out!
No kidding around, its serious business!
Pp. The links I dropped are for educational purpose, if they are edited sticky mail me.
I look forward to your feedback!
Joern Lucht, Berlin, Germany
Sites that are below PR4 you can't accurately do backlinks on.
if you mean by that (i might misunderstand it as i'm no native-speaker) that only links from PR4+ pages are being counted by google when they do their PR iterations i must strongly object. google just hides all PR0-PR3 pages on searches for "link:www.site.com". they do this (as well as not letting us combine the link: command with other commands) to obstruct the view on ther PR algo. it is to keep SEO people and competition off their secrets.
Here is what I intend to do until tomorrow... :)
- I will submit some internal pages of my site to DMOZ (i.e. www.mysite.com/wine/ to the Wine Category)
- I will provide our content (4,500 pages) to all of our 7 PR5 sites.
- We will NOT interlink the PR5 sites but we will provide some external links to our site from some selected upper pages of each PR5 site. The big content could help the PR of the 7 sites and they could also help the PR of our site.
We will see what happens then and I will certainly get back to you for feedback.
I will be waiting for your comments on these short term actions.
P.S. Many of you posted comments on the heavy workload that the content transfer would take. We use Coranto for our sites (free content manager software), so we just need 3 hours of work for each site... :)
Aris, submiting some internal pages of your site to the ODP may help considerably, and not just in Google. Unless you get listed in a very high level category, it would not help much in getting from a low PR7 to a PR8.
By not interlinking the PR5 sites you will not look so much like the many people who got PR0 a few months ago. Whether Google's human reviewers react kindly to interlinked duplicate sites may be a different matter. I don't see why providing the same content to your 7 sites will help you get more good links (and therefore PageRank) to them anyway.
There have been a few papers written by Web designers trying to understand Google; I have yet to read one that doesn't exaggerate the feedback effect of internal links. Internal links do count and there is some small amount of feedback, but adding extra pages doesn't help. Sorry, it's just not that easy:).
We have one PR6 site and 7 PR5 sites. I am going to assume that at this time they are not linked.
The biggest impact would be if you were to deep link to the lowest PR page in the main site. Do not link back.
You'd still have to link back to your other high-pr-goal pages from those pages to push that PR back up the chain.
From comments I've read so far, it is possible for a page to outrank the home page if it has more/better links than the homepage. So, IMO, send links where you need the PR the most, or where having it'll have the biggest affect.
