My question is this: Given that Google gave out the patent application for it's methods, how much of that has changed. One would think that with that kind of information every website would have a PR 10. Has Google changed that which it had originally stated in the application? Just curious......
your co-workers may be referring to 'Click Popularity' here. I don'y think there was any need of doing a test, as the result was onvious Some search engines consider click popularity as a factor of ranking webpages along with so many other factors. However, click popularity is not directly related to page views, but is related to some extent. It is related to the fact if someone clicked on your site from search results, and quickly came back to search the same keyword again, it means the user didn't find what he / she was looking for on your site. Here's an explanation. You can see a detailed explanation for Click Popularity here: http://www.spannerworks.com/click_popularity.0.html Several search engines use Click popularity algorithms for ranking webistes, others giving it less importance, and some of them not considering it at all. Also, IMO, opinion, Google pays some importance to click popularity, because if you use Google sitemaps, you will notice that Google shows stats for your site, and shows how many times it showed in SERPs for searched terms/keywords, and how many times your site was actually clicked through by users. And with Analytics becoming popular, Google might eventually give importance to pageviews and they might affect SEO for a website. With Google.. anything is possible
So you go to Google and do a search. You click on a link. If you are on that site for 5 seconds and return to Google then google recognizes that. If every visitor clicks back within seconds, its probabaly not a very good site. I think that Google sees that and if people hit that site and hardly return, then that means something, time spent on that site before hitting the back botton is recorded. Either your site is popular or isn't. and everything on the internet is tracked.
yes exactly, thats the concept. And search engines DO track this. And since they have so many users coming and searching a single term, and if the same pattern is followed for visiting a site by multiple users, search engines can decide which site has the info and which doesn't. This is also a good way to get rid of spammy sites or sites which are over optimized and might have achieved high rankings without actually providing useful information to the visitors.
it is also one of the easiest to manipulate. Google tracks what position you click on, but that is for relvancy, not for tracking sites.
well, i kept the experiment going and now "experiment 1" IS ranked first: g ranking for experiment so it looks like # of page views DOES effect rank. at least for now.
How do the search engines track the pageviews once you are in the site? There are a lot of assumptions made by the small breadth of this test. Thoughts?
Sorry but that test is not scientific at all. First off the page does NOT rank #1, it ranks #1 for the non www. but in the www. page #2 ranks at #1 . Sorry but if you are going to conduct tests you need to make sure that it is clinical. I am not having a pop, I am just saying that any test needs to be scientific in its approach, and the very fact that anyone could click on any page gives a MASSIVE element of doubt.
first off, who cares whether it ranks with or without www? second off, i have web stats Old Welsh guy so there isn't a MASSIVE element of doubt. third, i can do whatever i want, i do not "need to make sure that it is clinical". i never said it was perfect, but i think it gives a decent idea of whether or not # of page views effect rank. finally, the search ranking changed and now "experiment 1" does not rank first anyway. so it looks like the data center that was showing "experiment 1" ranked first is no longer live.
Whoa toofast, calm down, there is no need to attack me like that. I was not ahving a pop at you I was simply saying that you can not make sweeping statements like # of clicks affects rankings without clinical proof. It was not an attack on you or your experiment, I was simply sayingn that much more attention to detail needs to be made to enable rock solid proof.
I doubt there is traffic in the google algo. If it exists there will be a case of the rich getting richer and poor poorer (websites).
Thank you for performing experiment, but there are some issues with your experiment : 1. all visits came from one IP and since that it might not affect the SERP. 2. they might use different statistical approaches to detect interesting pages. OK, good page is the page : - which has large amount of repited visitors using bookmarks - visitors stays in average 5 secunds with 10 sec. variance (or something like that) - visitors who came to one page will more likely visit more pages on that site. So, that are some factors which might influence be included in statistical analysers for Toolbar data. In fact, I'm pretty assured that Google will use the Toolbar data for mentioned statistical calculation which will influence SERP rapidly, BUT I don't know when they will start to use data from Toolbar (or have they started). At the moment they might just collect data for later use, or doing nothing at all. They don't pay for spreading Google toolbar without purpose. OK, i visited experiment1 several times went back and forth.