Results 1 to 25 of 25
  1. #1

    web site traffic, PR and Google positions

    Does traffic increase have positive impact on web site positions in Google and PR? What are your opinions about this?

  2. #2
    Join Date
    Mar 2008
    Location
    Fort Thomas, Kentucky
    Posts
    269
    Not at all. Sure they monitor the clicks as much as we would on our own websites.
    Jaan Kanellis | iNET Interactive

  3. #3
    Join Date
    Mar 2004
    Location
    Pakistan
    Posts
    2,752
    Nopes not with Google. However, Yahoo, Lycos and AltaVista seems to be utilizing it to some level, but its role in their algorithms is not important.
    I'm Zafar Ahmed.
    I provide
    SEO Services & eMarketing consultancy
    I'll be glad to hear from you

  4. #4
    Join Date
    Mar 2008
    Location
    Fort Thomas, Kentucky
    Posts
    269
    Quote Originally Posted by Zafar Ahmed View Post
    However, Yahoo, Lycos and AltaVista seems to be utilizing it to some level, but its role in their algorithms is not important.
    Then why do you say here that Click Popularity is important to Yahoo?

    Quote Originally Posted by Zafar Ahmed
    Click Popularity:
    This must be something you have never heard of. But, this is a part of Yahoo’s algorithm. The more clicks your site gets on Yahoo Search Page, the more changes of your website getting on TOP 10. Yes, this is not part of Google’s algorithm and probably something new to you.
    Jaan Kanellis | iNET Interactive

  5. #5
    Join Date
    Mar 2004
    Location
    Pakistan
    Posts
    2,752
    Quote Originally Posted by incrediblehelp View Post
    The reason for posting the "click popularity" point in the last was to let people know that "click popularity" existed with yahoo or perhaps still exist.

    Are you saying there was no such thing as "click popularity"?, this may not be a part of Yahoo algorithm now but there surely was "click popularity" algo in few search engines including Yahoo.
    I'm Zafar Ahmed.
    I provide
    SEO Services & eMarketing consultancy
    I'll be glad to hear from you

  6. #6
    You said it does exist, now you say your article was writtin for the past and you do not know if it is still around or not?
    Technical Advisor for new A&E Series The Killing Season
    There are no random acts of violence
    Starts November 5th!

  7. #7
    Join Date
    Mar 2008
    Location
    Fort Thomas, Kentucky
    Posts
    269
    Quote Originally Posted by Zafar Ahmed View Post
    Are you saying there was no such thing as "click popularity"?, this may not be a part of Yahoo algorithm now but there surely was "click popularity" algo in few search engines including Yahoo.
    Well I don’t think anyone can say for sure, only employees of SE's can say for sur.. That is why I am shocked that said it did count at one time. As far as I can tell it doesn’t. I see some of my personal websites easily out-rank websites that I know get more traffic than me on all the major engines.
    Jaan Kanellis | iNET Interactive

  8. #8
    If memory serves click/clickback popularity was a part of the directory listings and never part of the serach results. Of course this goes back a few years so what I remember and reality may not match...I never claimed to be a guru but I am starting to feel like one
    Technical Advisor for new A&E Series The Killing Season
    There are no random acts of violence
    Starts November 5th!

  9. #9
    AHFB: at least most of what you say, that I have seen, holds water or at least is not utter conjecture.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  10. #10
    Join Date
    Jan 2004
    Location
    Oztrayla Mate!
    Posts
    583
    Quote Originally Posted by AHFB HTML View Post
    If memory serves click/clickback popularity was a part of the directory listings and never part of the serach results.
    Google does actually have a patent regarding the utilization of click data as a signal for scoring web documents. I have not seen any evidence to suggest this signal is being used, or if so it's currently "wound down" so the importance of this signal is minimal.

    We have done some tests with a good number of participants in attempt to promote one site and demote another by click activity, but couldn't obtain any conclusive evidence. The test simply put was, we ranked 2 sites in the top 10 for a low volume/low competition keyphrase where the results we quite stable. The participant queried the term, clicked on site A and clicked back to the SERP's within 10-15 seconds then clicked site B and browsed it for at least 5 minutes. This simulated site A not being relevant to the query and site B was.

    The thing is Google lodges a lot of "curve ball" patents where the patent idea is never intended for use. Also they lodge a lot that are crystal ball ideas, where they see a technology "may" be used in the future. This is also a step to prevent competition securing rights to the idea.

    But i have no doubt click data will become a more important signal in the future, and the reason is it's harder to "game" on a large scale unlike other factors such as backlinks and on-page attributes.

  11. #11
    Join Date
    Mar 2008
    Location
    Fort Thomas, Kentucky
    Posts
    269
    Quote Originally Posted by 1boss1 View Post
    But i have no doubt click data will become a more important signal in the future, and the reason is it's harder to "game" on a large scale unlike other factors such as backlinks and on-page attributes.
    Of course it will as personalized search grows.
    Jaan Kanellis | iNET Interactive

  12. #12
    Quote Originally Posted by incrediblehelp View Post
    Of course it will as personalized search grows.

    someone has their thinking cap on.

    I agree that as we see personalized search progress, we may see some new factors which just would not make sense in the general organics come into play.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  13. #13
    Quote Originally Posted by 1boss1 View Post
    The participant queried the term, clicked on site A and clicked back to the SERP's within 10-15 seconds then clicked site B and browsed it for at least 5 minutes. This simulated site A not being relevant to the query and site B was.
    The problem there is that google has no way of knowing how long a person sites on my page. They may never return to google for that phrase, instead opting to try their luck at msn or yahoo, or maybe they saw an ad on the first site they clicked that was what they wanted. That doesn't show relevancy, unless they have g analytics on everyones websites, which they are not even close to yet.

    Quote Originally Posted by 1boss1 View Post
    The thing is Google lodges a lot of "curve ball" patents where the patent idea is never intended for use. Also they lodge a lot that are crystal ball ideas, where they see a technology "may" be used in the future. This is also a step to prevent competition securing rights to the idea.
    Therein lies the rub. While patent watching can lead to some new theories, until they are proven (or disproven), they are simply theories and should never be put out as if they were fact where newbs might be mislead.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  14. #14
    Join Date
    Jan 2004
    Location
    Oztrayla Mate!
    Posts
    583
    Quote Originally Posted by nuclei View Post
    The problem there is that google has no way of knowing how long a person sites on my page.
    So nobody visits your page using the Google Toolbar, which is shipped pre-installed on Dell PC's?. It was estimated over 100 Million in 2006, and we are now in 2008. That's a pretty good data set right there without taking in to account pages with Adsense, Analytics and people logged in to their Google accounts for everything from Email to Gadgets and Docs. Then you have click through/click back rates from organic SERP's, Blogger, Feedburner as well as lumping the data on top of that from buyouts such as DoubleClick, YouTube, Performics and so on.

    If you think Google has no data about your page, that's amusing.

    Therein lies the rub. While patent watching can lead to some new theories, until they are proven (or disproven), they are simply theories and should never be put out as if they were fact where newbs might be mislead.
    That's why i always test theories, as demonstrated above. Actually i have been involved in several of these click data experiments over the years, the first was with some of the members of SEOChat back in late 2003ish i believe it was when the theory first arose.

  15. #15
    Quote Originally Posted by 1boss1 View Post
    So nobody visits your page using the Google Toolbar
    Funny you should mention that as I regularly check to see whats being sent by the toolbar. Sure it passes every url you visit, but it does not currently pass other possible factors such as time spent, referer, whether the browser was closed, etc.

    It's data being used as a possible ranking factor currently seems a tad in the realm of the twilight zone.


    Quote Originally Posted by 1boss1 View Post
    If you think Google has no data about your page, that's amusing.
    I never said any such thing. I simply said google had no way of tracking all occurences of user interaction and termination, which would be essential in realistically being used as ranking factors.

    Quote Originally Posted by 1boss1 View Post
    That's why i always test theories, as demonstrated above. Actually i have been involved in several of these click data experiments over the years, the first was with some of the members of SEOChat back in late 2003ish i believe it was when the theory first arose.
    Oddly enough I recall that test.
    Last edited by nuclei; 04-04-2008 at 03:33 AM.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  16. #16
    Join Date
    Jan 2004
    Location
    Oztrayla Mate!
    Posts
    583
    Quote Originally Posted by nuclei View Post
    Funny you should mention that as I regularly check to see whats being sent by the toolbar. Sure it passes every url you visit, but it does not currently pass other possible factors such as time spent, referer, whether the browser was closed, etc.

    It's data being used as a possible ranking factor currently seems a tad in the realm of the twilight zone.
    So it sends the URL you are on but not time spent on page. How do you work that out?

    If i'm on this page, Google knows the URL. If i visit your page Google knows the URL, and if i click the back button Google knows i was on your URL for "X" seconds before returning to this one. Seems like a time spent on page measure, and the clicking of a back button indicator which is what the click tracking patent is all about. Times this one action by several thousand and you start to see a pattern of people hitting the back button when landing on your page.

    Your whole rankings won't be based of this and this alone, it will be just one of a few hundred signals.

    The exact same data can be gathered from Adsense instead of the toolbar even though your page doesn't have it. If the referring page has Adsense, in this case WHT and i click through to your page and click back again in 10 seconds Google knows due to the add reload on my IP for this page.

    Hardly "Twilight Zone" stuff, i knocked up a script in a couple of hours that tracks my Adsense and supplies the same data.

    Anyhow, good luck with it.

  17. #17
    Quote Originally Posted by 1boss1 View Post
    If i'm on this page, Google knows the URL. If i visit your page Google knows the URL, and if i click the back button Google knows i was on your URL for "X" seconds before returning to this one.
    And if I close my browser? Reopen it and visit another page? Your talking about pages visited and not clicks when it comes to the toolbar data being sent out. Couple that to the fact that Matt has specifically stated that toolbar data is not used in the ranking algo "at this time"?

    You see this is where these arguements always fall short. What sounds logical or remotely possible is not always feasable in the terms of benefit from resources used. It takes the big G several days to calculate a toolbar pagerank export across a mere 20 billion or so web pages. Want to guesstimate how long it would take them to tabulate all these hits, clicks, backclicks, etc. from a whole lot more users from various sources thru numerous iterations just so they could then be run across the rankings of those 20 billion pages?

    Feasability. Always the keystone.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  18. #18
    Join Date
    Jan 2004
    Location
    Oztrayla Mate!
    Posts
    583
    Quote Originally Posted by nuclei View Post
    You see this is where these arguements always fall short. What sounds logical or remotely possible is not always feasable in the terms of benefit from resources used. It takes the big G several days to calculate a toolbar pagerank export across a mere 20 billion or so web pages. Want to guesstimate how long it would take them to tabulate all these hits, clicks, backclicks, etc. from a whole lot more users from various sources thru numerous iterations just so they could then be run across the rankings of those 20 billion pages?

    Feasability. Always the keystone.
    Was that post a joke?

    It's simple click logging, they already do it. Look at the attachment and here's searches i done in August 2006 complete with what i searched, URL's i clicked and time in between actions.

    So much for not feasible, this is just the simplistic data they give to us in your Google Web History that 99% of the population don't even know about or use. You think it's not feasable to tabulate this to incorporate in to web search which is the core of their business model.

    As for the resources used to do this this, just look at a shared host with Awstats running to log visitor interaction with webpages. Hardly processor bending stuff, i could write an app to log and process this data on my Nokia N70. I've already turned my phone in to a webserver so porting a stats program is doable all on 32MB RAM and 220Mhz processor while simultaneously running a webserver.

    I really hope you "intended" for you post to be comical, if so you succeeded.
    Attached Thumbnails Attached Thumbnails google.JPG  

  19. #19
    Quote Originally Posted by 1boss1 View Post
    So much for not feasible, this is just the simplistic data they give to us in your Google Web History that 99% of the population don't even know about or use. You think it's not feasable to tabulate this to incorporate in to web search which is the core of their business model.

    Are you seriously comparing the web statistics of a handful of sites to the billions and billions of clicks per billion pages in googles index???

    You are confusing simply pulling from a database data for a single person and their searches and somehow corelating that to running the number of iterations over what would amount to zillions of records that it would take to successfully be able to finally apply that data to rankings. That is just plain silly to compare. Foolish actually.

    Quote Originally Posted by 1boss1 View Post
    As for the resources used to do this this, just look at a shared host with Awstats running to log visitor interaction with webpages. Hardly processor bending stuff
    Good example, I run a shared server hosting quite a few people. Granted not a huge operation, but for this example it does not matter. Yet again, you are comparing the statistics of likely 50-150 websites on a server, yet it is suggested by the whm creators that awstats only be run once per day, per website, on shared servers to keep resource usage down. Now multiply those resources by billions which would be googles resource usage. Not feasable is a solid term that comes to mind for a simple tweak to rankings.

    Googles core business model? Their core model is advertisements, not search. They were just smart and didnt't pop the money stuff out til after they had the free part out there as #1, but do not confuse the two please.

    Assuming 5 iterations (I highly doubt it is that low) across all clicks on 20 billion pages per day, say an avg of 25 clicks per page to be on the low side of things, and you now have to multiply that by a months (or more) data to properly apply a filter to the results for any given search and you have a whole lot more resources than would ever make it feasable.
    William Cross
    Don Halbert *play site*
    william@seofox.com

  20. #20
    Join Date
    Jan 2004
    Location
    Oztrayla Mate!
    Posts
    583
    Ok i give up, you're completely correct. Its totally beyond Google to store timestamp values of time spent on a page and divide them to get an average. I mean if they go running queries like that their host might make them upgrade to the advanced hosting package.

    Oh yeah, my screenshot is fake too Google doesnt have my search queries, times, dates and URLs i clicked from 2006 because they would run out of disk space and this data has no importance to them what so ever.

    Thanks for the laughs "SEO Fox" you have a good one its been a blast.

  21. #21
    Quote Originally Posted by 1boss1 View Post
    Ok i give up, you're completely correct. Its totally beyond Google to store timestamp values of time spent on a page and divide them to get an average.
    Now you are being foolish again. If all it took was counting clicks and averaging them, not taking into account may other factors such as cickbots, which clicks were for which keywords, which clicks came from personalized search, which clicks were a result of skewed results from having your google account active while searching, then sure, it would not be that difficult. But then again, maybe you just did not think that deeply into the idea of what it would take to actually have it mean anything. I have, years ago, and I have even kept up with the patents.

    Quote Originally Posted by 1boss1 View Post
    Oh yeah, my screenshot is fake too Google doesnt have my search queries, times, dates and URLs i clicked from 2006 because they would run out of disk space and this data has no importance to them what so ever.
    As explained above, you are comparing the resources used to store and pull up a single users history to that particular user, to the resources it takes to run somewhere in the area of 16 zillion iterations to accurately use the data as a filter. I would have thought you were brighter than that from most of your posts.

    Quote Originally Posted by 1boss1 View Post
    Thanks for the laughs "SEO Fox" you have a good one its been a blast.
    Why do people always resort to this when they simply can not provide logical arguements?
    William Cross
    Don Halbert *play site*
    william@seofox.com

  22. #22
    Join Date
    Mar 2008
    Location
    SEO cyberspace
    Posts
    423
    I believe that Google do not store the web history automatically just because you have the tool bar installed, you have to sign up for the service and even at that you have the option to limit your history to searches performed, NOT every site you visit and in addition to that you have to have a Google account.

    These factors IMO result in far fewer data sets than there are toolbars, and the stated purpose is for use in personalized search.
    Web History records information about the web pages you visit and your activity on Google, including your search queries, the results you click on, and the date and time of your searches in order to improve your search experience and display your web activity. Over time, the service may also use additional information about your activity on Google or other information you provide us in order to deliver a more personalized experience.
    This IMO does not seem like any sort of a foundation for algo modification for all google searchers.
    I plan to live forever - so far so good
    Expert SEO |Sash Windows London

  23. #23
    Ok guys, I have another question: do search engines take into account such factors as how much time do users spend on visiting web site, how often the pages are updated, etc?

  24. #24
    Pages updated, yes...time online no
    Technical Advisor for new A&E Series The Killing Season
    There are no random acts of violence
    Starts November 5th!

  25. #25
    page freshness can at times tip the scale on similarly placed listings IMO, however time spent on a page, no.
    William Cross
    Don Halbert *play site*
    william@seofox.com

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •