Forum Moderators: Robert Charlton & goodroi
Please bear with yet another newbie question :) I've been reading quite a lot about SEO these past few months and I decided to perform an experiment. I'd appreciate some comments on my findings.
I chose a keyword for which the No1 site is a PR4 site with 400 pages indexed and 17,000 backlinks (according to MSN).
I optimized my own site for that very same keyword and I got about 250 pages indexed. My site is now a PR3 but with only about 800 backlinks. My keyword density is about 6.5% on the first page (while that other site has about 5%) and the number of my pages that actually contain the targeted keyword is much higher than that of the other site.
However, it's No.1 and I'm nowhere to be found in SERPS.
Now, based on all that and all that I've read, am I correct to assume:
- PR is not necessary a measure of backlinks. Even if my site was PR4 with say, 1000 backlinks, the PR4 with the 17,000 backlinks would still rank higher.
- The site's age is really really important. It's pretty much impossible for a new site (mine is 3 months old) to rank in the top 10 for a competitive keyword (unless it's got a lot more pages than the current top 10 sites)
- If keyword density, and keyword frequency were really that important regarding relevance, I would at least be somewhere to be found in the SERPs.
Another thing that I tried was posting an article, about 1000 words, with the keyword only mentioned in the title and maybe 3-4 times in the article body. Sure enough, the article page on ezinearticles.com does show in the top 10!
So, I'm trying to figure out where the "PR is no longer important" thing I've been reading so much about fits in.
Please note that I'm not doing this to be the No.1 result in one afternoon and make thousands of $ while sleeping (this is for my other site:), but I'm really trying to learn about SEs and maybe, although a newbie, my experimental site can help in discovering a bit more about how these blasted things work! :)
Thanks for your time and sorry for the long post.
Page Rank is not just dependent on the number of backlinks. Other factors are the PR of the linking page and how many total links arre on that page. Also, some pages are now tagged by Google not to pass on any PR "vote".
Also, 3 months is a short time. Your site may well be experiencing the Google Sandbox effect.
PR is absolutely useless from a ranking point of view.
Read up in the forum about:
Sandbox
Link dampening
Random Google 'bad days'
Welcome to the world of SEO!
PR is absolutely useless from a ranking point of view.
In my view, that is a bit overstated. PR is one of many factors that DO influence final ranking on search queries. Yes, a PR zero page can outrank a PR6 page for this or that search because of all the other factors, including the search phrase itself -- the PR of a page is independent of any connection to keywords or content. But that doesn't mean PR isn't in the mix, it definitely is.
Also some points to consider when optimising your content are things like: sylable counts, number of sentences, how many words per sentence you average, Fog Reading Grade Levels, Flesch Levels, Kincaid Grade Levels ... all of which can be calculated and evaluated using tools online.
These all determine how easy google think your page is to understand and therefore how wide your audience should be.
All the best
Col :-)
PR is absolutely useless from a ranking point of view.
That's quite a declaration. On what is it based?
Our pages always rank higher when they have the correct PR.
In one case a PR6 page that ranks very well was effected by cannonical problems and went PR0. It no longer ranked for the chosen keywords even though it was in the index.
When the PR returned so did the ranking.
What do you mean?
In particular if a website fell in penalty the page rank could have a negative influence? the website could lose pr?
Thank you!
However, I can't resist trying to make yet another naive analysis about the sandbox :)
Suppose that you wanted to write an algorithm to sort 100 mil elements, so that you can produce result pages with the most relevant element first. If you did that by assigning a relevancy or ranking number (fictional factor, nothing to do with pagerank) to each item, and that number had a range of 1-100 then with 100 mil items, the competitions would really be in the decimals! So, item 99.09 would be pretty low on the sorted list, if all other spots up to 100.00 were taken! If you also subtract a dampening factor of say 1 (or multiply it by 0.9) then it's even worse!
If we were talking about SERPs, then this could help explain two things:
1) For competitive keywords it's more likely to have a lot of well optimized pages and small granularity, so the competition is really is in the decimals of the fictional ranking number.
2) For non-competitive keywords you have less highly optimized pages and higher granularity so if the first site has a ranking number of 98 and the second a ranking number of 92 you can really afford loosing some points and even outrank the number two result. This explains why some people claim to be able to avoid the sandbox for non-competitive keywords.
What do you think?
About pagerank, do you think that number and quality of backlinks are evaluated separately and not accurately represented in the PR?
Also, I'd like to thank the person who provided the info about determining reader's level for a page. Never would have thought of that!
Take care!
The pages are all unique, it has well wrote unique meta tags, uses no flash
That makes me believe that its all about unique content!
That makes me believe that its all about unique content!
I agree, but it seems is rather new and began from Big Daddy.
They show rather Daddies, i.e. original producers, and the Daddies are big (high in SERP)
Vadim.
Page Rank is not just dependent on the number of backlinks. Other factors are the PR of the linking page and how many total links arre on that page. Also, some pages are now tagged by Google not to pass on any PR "vote".Also, 3 months is a short time. Your site may well be experiencing the Google Sandbox effect.
YES!
It has been 8 years and besides some like Ted, it seems - very few people get PageRank. It really isn't a hard concept.
PR = PR of other pages linking to you / number of links on that page - dampening factor.
Now even without a sandbox - it can take months to see it, but you still might have it. Matt & Co don't like to update the PR on the toolbar as oftenn as the do internally.
Google has experimented and does in some cases (as Ted mentioned) either not pass on pagerank from certain pages - or passes less than they should.
I spent probably a year of my life researching PageRank. There are probably people out there that know more about it than I do now - as I don't pay as much attention to it now as I did then.
PageRank is important. However, it is not as important as it was.
In the old days - I could make Satan out rank anyone for the term "easter bunny" if he/she had more PageRank than you.
I had blank and error pages rank #1 for competitive terms if I had enough PageRank.
However, Google has now moved into what I view as the third generation of ranking factors: Signals
First there was IR, Then PR, now Signals. If you don't have either a certain amount of good signals or have too many bad signals - and this is all relative - PR isn't going to help you.
Although Google has "over 100 ranking factors" and now has X number of signals - they still have PR. They don't need to change PR - they change the weighting of it in their algo - and tweak a couple things to it here and there.
I know of two specific changes they made to the PR calculation - I am sure there were others, but the two I know of were relatively minor and would not effect most people - and although google never mentioned that they changed them - they told you about them in different ways.
PageRank is important. However, it is not as important as it was.In the old days - I could make Satan out rank anyone for the term "easter bunny" if he/she had more PageRank than you.
I had blank and error pages rank #1 for competitive terms if I had enough PageRank.
However, Google has now moved into what I view as the third generation of ranking factors: Signals
First there was IR, Then PR, now Signals. If you don't have either a certain amount of good signals or have too many bad signals - and this is all relative - PR isn't going to help you.
I tend to strongly agree with these statements.
However, ALL THINGS BEING EQUAL - you will stil lose out to higher pagerank.
That is, in the case of 2 sites optimized for the same type keywords, neither penalized, the higher PR site will still come ahead of the second in SERPs.
However, as pointed out PR is not simply number of links. In fact for a medium size site, I would think 800 inbounds probably means a lot of cr@p that isn't helping and may be hurting. There are a lot of sites with 10 or less backlinks with page 1 SERPs and with an "appropriate" PR (e.g. for a small-medium sized niche site, PR5 is very acceptable).
In this case, the visible pagerank is only the effect of having links to your site.
I have a new site up, less than 200 pages, no backlinks and it is ranking top 10 in the serps for some extremely competitive key words.
no backlinks but ranking top 10 for extremely competitive kw? You are talking about Google, right? You must be lucky.
Unique content is not enough to rank high. There are too many sites with unique content (not as many as junk sites, but still a lot) to put them all on the first page. There is a clear correlation between the PR and the rank.
It does matter.
um, no, not in google, i am not buying it.
Also, the competitor with the 14,000 links in msn, you can jsut about bet your last 2bob, that these are links from site that have scraped dmoz, or are otherwise just plain crap.
14,000 inbound links for pr4?
ONe of my sites is pr6 with 190 inbounds.
My advice?
Add a 20-30 links a month and concentrate on your content and onsite anchor text
Take care!
