View Full Version : Google results, time, and psychology
Aerik
07-23-2005, 02:37 PM
Does anyone know if there have been studies to corrlelate the age of pages and their pagerank? What I'm getting at is that the longer a page is on the net, the longer it will have had for people to link to it. I wonder if Google tries to figure out the effects of that, or if it just results in better weightings for established pages.
If it results in better weightings for established pages, you can look at that two ways: you can say that it's appropriate because the pages are, after all, established, or you can say that this unfairly penalizes new content. Many folks (non experts) will really interpret Pagerank to literally be a measure of a pages relevance - which of course it is not. An article in Wired magazine just had a great definition of Pagerank - that it is basically the same as measure the citations of an academic paper. As many have experienced, poorly formed searches will sometimes result in very irrelevent search results - even the top results.
I was also wondering about the psychology of linking - as pages become more and more established, are people less likely to link to them? Are tech saavy people more likely to link to pages that are not established, knowing that they need the boost?
All of these questions came about because I have been thinking a great deal about the usefulness of search results and what it means to the "long tail" of internet contributors. The average blogger is not ever going to get top search engine results. Some guy who publishes a fantastic tutorial on how to repair old chevy small block engines may get good rankings, but unless he does some good SEO, it will take a very long time for him to organically develop the links to build his PR.
So does that mean SEO is becoming a necessity to anyone who wants to get found? I think it does - but it also means that there are opportunities for alternate schemes (I've also been researching, separately, whether people still see any direct value - not for SEO - in directories).
I think some interesting discussion could spring out of this - or perhaps already has?
Warm Regards,
Aerik
Relevancy
07-27-2005, 02:37 AM
Does anyone know if there have been studies to corrlelate the age of pages and their pagerank? What I'm getting at is that the longer a page is on the net, the longer it will have had for people to link to it. I wonder if Google tries to figure out the effects of that, or if it just results in better weightings for established pages.
Well there is weight with age of domain more then age of individual pages. A domain that has been hosted for 5 years will most defiantly out rank a site that is 10 months old. There are the other factors such as links and site development, but if you were to start with similar sites at the same time but one is 4 years older you will find much faster success. To find the archived age of a domain use: http://www.waybackmachine.org/. This is better then looking at the register info because that will only tell you how long it has been registered for and not if it was ever hosted or crawled.
I was also wondering about the psychology of linking - as pages become more and more established, are people less likely to link to them? Are tech saavy people more likely to link to pages that are not established, knowing that they need the boost?
People link to the best resource period. And the best resource usually ranks better.
So does that mean SEO is becoming a necessity to anyone who wants to get found? I think it does - but it also means that there are opportunities for alternate schemes (I've also been researching, separately, whether people still see any direct value - not for SEO - in directories).
SEO in its true form is not a necessity for getting found… site development and becoming an authority on your subject matter will bring results and traffic. Page clarification and site theming is needed to clearly represent your site's worth.
seobook
07-27-2005, 09:37 AM
Does anyone know if there have been studies to corrlelate the age of pages and their pagerank? What I'm getting at is that the longer a page is on the net, the longer it will have had for people to link to it. I wonder if Google tries to figure out the effects of that, or if it just results in better weightings for established pages.
there is first mover advantage, but in competitive fields that can erode. look how Google came from nowhere after none of the other search engines wanted to buy their PageRank patent.
If it results in better weightings for established pages, you can look at that two ways: you can say that it's appropriate because the pages are, after all, established, or you can say that this unfairly penalizes new content. Many folks (non experts) will really interpret Pagerank to literally be a measure of a pages relevance - which of course it is not. An article in Wired magazine just had a great definition of Pagerank - that it is basically the same as measure the citations of an academic paper. As many have experienced, poorly formed searches will sometimes result in very irrelevent search results - even the top results.
Google's algorithm is much more advanced than just using PageRank. they give some noticable boost to some new documents, but if they are competing for hypercompetitive terms they will need a ton of linkage data to try to compete with older established sites.
you also have to look at it on a business front as well. the greater the cost to manipulate the results (cost = time + money) the more likely a site is to need to buy AdWords to try to get exposure on Google.
I was also wondering about the psychology of linking - as pages become more and more established, are people less likely to link to them? Are tech saavy people more likely to link to pages that are not established, knowing that they need the boost?
linking is self reinforcing. the top sites continue to get more exposure
http://www.e-marketing-news.co.uk/Oct04/RichLinking.html
although search engines can program some types of randomization into their algorithms to try to give newer sites a shot.
All of these questions came about because I have been thinking a great deal about the usefulness of search results and what it means to the "long tail" of internet contributors. The average blogger is not ever going to get top search engine results.
i have a personalish blog which has few links that ranks for a wide variety of search terms and gets hundreds of visitors a day with only about 400 posts on that site. people search for lots of random stuff. consider that:
around 50% of all search queries are unique
Google PDF (http://www.research.ibm.com/haifa/Workshops/searchandcollaboration2004/papers/haifa.pdf)
there are hundreds of millions of searches per day
there are less than 10 billion documents in the search indexes
there are at least 10 search results on the page
Some guy who publishes a fantastic tutorial on how to repair old chevy small block engines may get good rankings, but unless he does some good SEO, it will take a very long time for him to organically develop the links to build his PR.
well typically if someone publishes a ground breaking earth shattering document they have some background and friends in the field in which they are experts in.
it should take time to build brand, credibility, and linkage data. if your name was great in your industry then some of that carries over to the web almost right away as people say xyz from pqz just got online.
So does that mean SEO is becoming a necessity to anyone who wants to get found? I think it does - but it also means that there are opportunities for alternate schemes (I've also been researching, separately, whether people still see any direct value - not for SEO - in directories).
I think some interesting discussion could spring out of this - or perhaps already has?
in the past people needed ways to get found and published. they still do. as time passes there are more competing channels and more noise, but also more network users and a broader variety of ways people can find you.
the problem with directories are
scalability
monetization
inherant human bias