jump to last post 1-29 of 29 discussions (148 posts)

Plunges caused by Google restructuring subdomain linking power?

  1. PaulGoodman67 profile image89
    PaulGoodman67posted 5 years ago

    Article for discussion.  I know that this recent development has already been mentioned by some hubbers in forums.  But I am now wondering if this might be the main reason why we are seeing the current traffic plunges for many hubbers?  See link below:

    http://www.kingpin-seo.co.uk/webmaster- … wer/011230

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      I read it twice and still didn't understand what the author was trying to say.

      He seemed to go on about links from the root domain (Hubpages) to the subdomains having less value, but as far as I can see there are next to none from Hp to me, instead the links come from other subdomains.

      However, as the majority of my hubs are on a 301 re-direct, they may well have root domain linking in place, and I don't know if this carried over.

      Do my newer hubs written since the subdomain change came in, have a 301 too? They can't have.

      If someone understands all of this, perhaps they could explain it all a bit better?

      1. wilderness profile image94
        wildernessposted 5 years ago in reply to this

        You will likely have a few links from HP into your subdomain.  Featured hubbers, featured hubs, maybe the hub hopper - that sort of thing.  I once found one of mine on a section of HP dedicated to education that I didn't even know existed.

        Your page will also have links back to HP - the learning center, about us, etc.

        As you say, though, there aren't a lot of them.

      2. toygurus profile image60
        toygurusposted 5 years ago in reply to this

        If I am understanding the 301 redirect, I think even the newer hubs have them.
        When you go to
        http://hubpages.com/hub/my-hub-title
        it redirects to
        http://mysubdomain.hubpages.com/hub/my-hub-title

  2. wilderness profile image94
    wildernessposted 5 years ago

    I'm not sure that the topic here (links between a subdomain and the main domain) are particularly applicable.  Yes, there are a few links provided between HP and each subdomain, but not a great many.  Similarly, there are a few links from the sub to the main domain.

    More important, it seems to me, are the thousands of links between subdomains.  Each hub has a set of "related hubs" that link to other subdomains and well as an occasional in-text link that hubbers provide via the link suggestion tool.

    Each hub has a lot of (now external) links and it is not uncommon to see a hundred or more coming from other subdomains into one of my hubs.  Add in the fact that these links appeared overnight, so to speak, and it is a recipe for problems.

  3. Danette Watt profile image88
    Danette Wattposted 5 years ago

    what is a 301 redirect?

    1. wilderness profile image94
      wildernessposted 5 years ago in reply to this

      When you went to a subdomain the URL, your address if you will, changed for each hub.

      a 301 redirect is a "street sign" that HP put up for people looking for your old address; it sends the searcher on to the new, proper, address.

      1. Danette Watt profile image88
        Danette Wattposted 5 years ago in reply to this

        thanks

  4. toygurus profile image60
    toygurusposted 5 years ago

    According to Google the change in the way it is handling subdomains is only reflective in the web master tool and should not affect rankings. So there must be some other reason some hubpage account are not getting any traffic. See the article here. http://www.seroundtable.com/google-subd … 13968.html

    1. PaulGoodman67 profile image89
      PaulGoodman67posted 5 years ago in reply to this

      Thanks for posting, toygurus.  Intersting article.  There is clearly something drastic going on with the algorithm and HP.  Google aren't going to tell us exactly what they are doing, so I find the temptation to speculate to be overwhelming! haha!

      HubPages' strategy of using links to boost the SEO of the hubs and site was one of the main things that p*ssed off Google the first time around (because they saw it as boosting bad hubs as well as good) and panda was in part an attempt by Google to cancel out the boost that HP gave to hubs this way.  (That's my understanding anyway).

      Now HP has got around panda and found a different way of getting high rankings (read: high traffic) it seems reasonable to speculate that links are involved on some level.  It's idle speculation, and I know it's a complicated technical area and certainly outside of my solid grasp.  My previous technical knowledge is in things like programming and of  limited use with something like this.

  5. Marye Audet profile image91
    Marye Audetposted 5 years ago

    I have not seen a drop, in fact my traffic is at it's highest, non-holiday point ever. I think if you have developed a strong online presence in a variety of venues then this will matter very little. I believe that the point now is that Google is looking for an author's authority  more than the site that his work is found on.

    1. janderson99 profile image87
      janderson99posted 5 years ago in reply to this

      This has been discussed in other threads. It is informative to look at the links listed in Webmaster Central. I have 30,000 links !!! from other HP subdomains, most generated by 'related hubs'. Some of these are still in the old 'folder' format, before subdomains. These links are listed as external because the subdomain is owned(??) or assigned to the author. The internal links are simply a list of article pages. The links from subdomains must, in my opinion, be given less weight than external links. The churning through the linking structure is likely to be a cause for volatility. This takes time as it depends on when each page is revisited by the bot (every few days). Each time the reassessment is done the ranking of the links will have changed because of the subdomain shift. So this will take weeks to settle down and is likely to be volatile and may be seen as picking on some poor hubbers for a while.

      I personally don't like the 'related hubs' because many of them are totally unrelated to my pages and many of them are of low quality. I would like to be able to choose my relatives. Why the apparently random plunge and surge? I suspect that it is related to links and the PR of each subdomain. I believe that external links are extremely important - not the social links which are just fluff. I am lucky to be a member of the steady surge group and I think this is related to my external links - but one example never proves a theory. I don't believe that SEO is dead. Google's rankings still depend on its historic algorithm which relies on the number and quality of the backlinks. Panda is essentially a manual penalty which essentially does something akin to : "We don't like Hubpages so lets halve the PR of all its good articles and reduce the PR of the poor quality ones to only 10%". So it has got a lot tougher and you need better quality and more links to boost your PR. The steady rise of HP on http://www.quantcast.com/hubpages.com shows that it takes time. Remember for every winner there's a loser. The good and well linked articles rise the the top and claw their way over the poor hubs that have been downgraded. Its the survival of the fittest (dog eat dog) with fitness determined by Googlebots and Panda's qualty assessment and backlinks, backlinks, backlinks.
      Just my opinion of course. If I had plunged (or am about to), I would have a totally different view and 'I don't throw stones' !!

      http://s3.hubimg.com/u/5486094_f248.jpg

      1. thisisoli profile image64
        thisisoliposted 5 years ago in reply to this

        PR is different to page/domain trust.  PR is a simple algorithm that is not affected by any algorithmic change other than the buffer effect. It is a predictable calculation which is based only on the number of links, not on teh quality of links.

        You also need to change your quantcast graph to 1 year isntead of 1 month, we are not back yet tongue

        Panda is also an algorithm not a manual adjustment, it is based of indicators which were brought together through a human review panel, but Panda is an algorithm which judges a wide range of 'quality' factors and then applies a value to the main serps formula.

        However, Link quality is important, high quality links are definitely worth their weight in gold, and continue to be worth a lot after Panda.

        1. Richieb799 profile image64
          Richieb799posted 5 years ago in reply to this

          Hi Oli, I think backlinks are more important than ever now, We are essentially like dozens of other web 2.0 sub domain sites that rank on their own merit and don't benefit from the rest of the site content.

        2. janderson99 profile image87
          janderson99posted 5 years ago in reply to this

          I beg to differ on:
          "which is based only on the number of links, not on teh quality of links"
          and
          "Panda is also an algorithm not a manual adjustment"
          But have no time for a debate.
          Give HP credit where credit is due. The recovery has been outstanding and its tactics courageous. I have reaped the benefits and I'm back publishing hubs like crazy. I received my highest HP Adsense payout ever, yesterday in two years (400% above my previous best). So keep on Dancing Baby, Till the Night is Gone!

          1. thisisoli profile image64
            thisisoliposted 5 years ago in reply to this

            You can find the page rank algorithm both on wikipedia and Googles own blog, it works on the number of links, the only value that is passed from these links is the page rank of the source site.

            The quality factor of links, and other link values are important, but it is in no way related to PR. Confusing the two is like confusing a cogwheel with a watch.

            The Google panda update is an algorithm that is run approximately once a month, it is based on quality factors, and you can view what factors it was based on in the GWT blog. 

            The Panda algorithm is not a Google engineer sat at a desk saying 'This month I think I'm going to penalize Hubpages this much!'

            You can not want to debate as much as you want, but the facts are very clear on these two topics.

            @Paul - there are risers and fallers but for teh most part it seems to be sharp and cyclical changes.  I have risen and dropped several times now, and think it is worth waiting for stability before confirming which direction Hubpages is heading :p

            1. wilderness profile image94
              wildernessposted 5 years ago in reply to this

              I agree with what you are saying here, Oli - it is what I've been thinking for some time now.  Our surge is due to the sub to sub links provided by HP and the surge/plunge cycle is likely due to the bots seeing those links at different times.  In addition those links come and go (I don't know how HP assigns them, but they DO change).

              I would hate to see those links go the way of the dinosaur - would we then all go to basically pre-panda levels and steady out there?  Quite likely, I think, and I prefer what I have.  Of course I'm a "permanent" surger and that's easy for me to say.

              Or should we limit the number of hubs in a subdomain and thereby limit the number of these "external" links that can happen?  Most, but not all, of the cyclic respondents in Marisas hub have quite a few hubs.  It could well be correlated with subject matter as well - maybe HP changes those links more often on popular categories.

              Lots of unknowns yet, but I believe this is at least a good part of the problem you and others are seeing.  I'm also not totally convinced the radical ups and downs will ever stop.

              1. PaulGoodman67 profile image89
                PaulGoodman67posted 5 years ago in reply to this

                @Oli  The Google page rank/ algorithm is one of the few things I feel I have some grasp on, having studied probablity distributions. 

                @wilderness  I am increasingly beginning to think that Google will keep shifting the goalposts until they're put "content farms" in their place.  Therefore pre-panda levels of traffic would be unacceptably high to them in the medium to long term.

                1. wilderness profile image94
                  wildernessposted 5 years ago in reply to this

                  I really don't think this current roller coaster is the result of any overt google action.  It is rather the unintended and surprising result of the combination of our switch to subdomains, all those thousands of backlinks and the google algorithm.

                  I would be interested in hearing what someone getting large numbers of bing or yahoo search results are seeing from those traffic sources.  Surely there are a few of them around?

                  1. thisisoli profile image64
                    thisisoliposted 5 years ago in reply to this

                    Wessman - not necessarily, all old URL's are redirected. This is not 100% reliable but it is not worth duplicating everything.

                    As to the shift in backlinks, it is probably one of many things affecting us right now, until we have a glimmering of stability it will be hard to judge what's causing what.

                2. Greekgeek profile image98
                  Greekgeekposted 5 years ago in reply to this

                  This. I think from now on, any site that is an open-publishing platform is going to have a harder time outranking ones with editorial boards. (Although I can't tell that for sure, since eHow doesn't report to Quantcast directly.)

                  Panda is a separate algorithm run less frequently than the daily one. Every now and then -- when someone pushes the Panda button manually -- Panda crawls domains and gives all of them a quality rating based on multiple factors. We don't know exactly what those factors are, but since Panda is all about trying to determine the quality content / spam & scraped content ratio across a site, I'm fairly sure it tries to determine the extent to which content is vetted: 
                  -- can users flag content? (HP = yes)
                  -- are there spam filters which don't permit certain kinds of content? (HP = yes)
                  -- what kinds of content do spam filters block, and how much gets through?
                  -- can content ONLY be published once it's vetted by moderators or editors?  (HP = no, eHow = yes)
                  Panda then assigns a "Panda Rating" (whatever you call it) to each domain, and this rating is then one of the factors -- a strong one -- in the 200+ factors Google uses to determine how to list pages in SERPs.

                  From what I can see, Hubpages is now ranking similarly to, or a little better than, similar sites (ezinearticles, associated content, squidoo, suite101, gather). It might not be realistic to expect HP to get double traffic to all of those as it did before. Unless its content is really, truly, across the board, at least twice as good as content on all those sites?

          2. PaulGoodman67 profile image89
            PaulGoodman67posted 5 years ago in reply to this

            '"Panda is also an algorithm not a manual adjustment"
            But have no time for a debate.'

            Yes, I think you are wise to walk away.  ;-)

          3. janderson99 profile image87
            janderson99posted 5 years ago in reply to this

            Keep on Dancin' Baby - with apologees to Abbbbbaaaa

            I was at a party, this guy called Panda said to me
            "Something bad is happening, Don't you all agree?
            Google cares for nothing, no one like you and me
            Evil times are coming, traffics crashin', just you wait and see"
            I said, "Who are you to talk about Traffic set to doom?"
            He got kinda scary as he looked around the room
            He said, "I work for Google, I know the formula ain't great !"
            I said, "I just can't believe it, why don't my hubs rate?
            Panda, can you tell me what is right and what is wrong?"
            He said, "Keep on rocking baby, 'til the night is gone"

            On and on and on
            Keep on rocking baby
            'Til the night is gone
            On and on and on
            'Til the night is gone
            On and on and on
            Keep on rocking baby
            'Til the night is gone
            On and on and on
            'Til the night is gone

            Standing up is scary if your Hubs have started to fall
            Like Humpty Dumpty, its like falling off the wall
            Don't ask 'cause Google will never say what's going on
            Its all a secret so just keep writing and moving all along
            Gotta keep on rockin baby till the night is gone

        3. Marisa Wright profile image93
          Marisa Wrightposted 5 years ago in reply to this

          Tell me about it.  I have a PR3 site that got dumped in the last Panda update.  I obviously don't understand PR properly, it would never have occurred to me that my site could be simultaneously a PR3 and in the sandbox.

          1. thisisoli profile image64
            thisisoliposted 5 years ago in reply to this

            PR basically measures link numbers, not link quality.

            1. janderson99 profile image87
              janderson99posted 5 years ago in reply to this

              It depends on the PR of the links - to me that means link quality. Sure the SERP uses other stuff but the PR(quality) of the links matters, not just the number of links. Other wise why do people pay to get a link from a PR5 site?

              1. thisisoli profile image64
                thisisoliposted 5 years ago in reply to this

                And the PR of those links is based on the PR of links pointing to that link. PR is no indication of quality, only an indication of the number of links pointing to links (with a value based on the number of links pointing to the source link, and a buffer value thrown in).

                Even google has said in black and white that the importance of PR has been heavily reduced, because they have much better link ranking factors now that can measure the quality and relevance of a link among other things.

                People pay for a link from a PR5 site because there are a lot of people who don't know what they are doing smile

                There is a guy who currently has a network of shit sites, none of which rank in Google, most of which are PR6 - PR7, he earns close to £20k a month off these, just selling links.

              2. Greekgeek profile image98
                Greekgeekposted 5 years ago in reply to this

                <tangent>Why anyone would pay for pagerank is a mystery to me.

                (a) Pagerank is just one of over 200+ ranking factors Google uses to decide where to list a site in SERPs. Its significance has dwindled year by year. At this point the anchor text in your links is almost more important than pagerank.

                (b) Google's algorithm does everything it possibly can to detect paid links, and will penalize them unless they are nofollowed.

                (c) Google doesn't actually give away the Pagerank algorithm it's using now, and all those silly tools out there are based on outdated reports on how the pagerank algorithm USED to work when it was first patented/announced. We have no idea what Google's actual PR calculation for each page is; even the Google Toolbar is merely an estimate.

                So paying for high PR is liable to get a site dropped in the SERPs. </tangent>

      2. Silver Rose profile image77
        Silver Roseposted 5 years ago in reply to this

        Ummm... Hubpages isn't quite back - you are looking at the 6 month graph, which starts AFTER the Panda plunge.

        Here's the graph you want:

        http://www.quantcast.com/profile/trafficGraph?wunit=wd%3Acom.hubpages&amp;drg=&amp;dty=pp&amp;gl=1yr&amp;reachType=period&amp;dtr=dd&amp;width=720&amp;country=UK&amp;ggt=large&amp;showDeleteButtons=true&amp;v=-467210935

        1. PaulGoodman67 profile image89
          PaulGoodman67posted 5 years ago in reply to this

          Plus the graph suggests that the "recovery" might now be petering out (as more hubbers plunge?)

    2. Wesman Todd Shaw profile image92
      Wesman Todd Shawposted 5 years ago in reply to this

      That's what I thought until yesterday when the crash that others were talking about that I did not understand made it's way to my subdomain.

    3. Dorsi profile image92
      Dorsiposted 5 years ago in reply to this

      I agree. This is my "gut" feeling too. My traffic has quaudrupled from a week ago. Normally 1,500 views pre-panda then to 800 after Panda. Now close to 4,000 a day. And hubs I wrote regardless of good SEO tactics are rising too. The only thing I can determine is that my online presence has been rated in some way with Googles algorithm and I've got the thumbs up.

      As far as this article goes, alot of it is foreign to me. SEO has not been my strong point. Although I have gotten much better at writing with keywords and traffic in mind. Now when I write I do take that into consideration.

  6. thisisoli profile image64
    thisisoliposted 5 years ago

    The article was not particularly good, for starters Google have not announced they are viewing sub domain links differently, only that they are reporting them differently. The GWT Team is only loosely related to the SQ team.

    WHile it does indicate there may be a shift in how subdomains are viewed by the Google algorithm, there needs to be a lot of testing done before anyone can really classify whats been changed, if there are variables, etc.

    In relation to Hubpages traffic, mine has been swinging wildly in any case, with 90% drops and rises days apart.  This rollercoaster ride would be impossible to pinpoint without a full statistical analysis, but considering this has been going on for weeks now, I doubt it is all down to one Google change.

    1. PaulGoodman67 profile image89
      PaulGoodman67posted 5 years ago in reply to this

      Well, it was just a topic for debate.  I appreciate the issues around it are a bit too technical and specialized for a general discussion forum, but I thought I'd post it anyway.

  7. thisisoli profile image64
    thisisoliposted 5 years ago

    heh, I couldn't be bothered to take the snapshot of the 1 year graph but it looks like you were :p

    1. Silver Rose profile image77
      Silver Roseposted 5 years ago in reply to this

      I just right clicked on the image on quantcast to get the image url and hotlinked it - taking snapshots is too much trouble!

      1. thisisoli profile image64
        thisisoliposted 5 years ago in reply to this

        Damn, gonna have to remember that!

  8. Anna Marie Bowman profile image90
    Anna Marie Bowmanposted 5 years ago

    I don't pretend to understand all of what is going on with Google, and with all the new subdomains, but I think if you write quality Hubs, you shouldn't have a problem.  I have seen a jump in traffic that has stayed steady for over a week now.  I personally think that whatever is going on is only going to work to weed out the bad Hubs, and bad Hubbers, and improve the quality of HubPages as a whole.

  9. Wesman Todd Shaw profile image92
    Wesman Todd Shawposted 5 years ago

    Are you saying we should re backlink things?

  10. thisisoli profile image64
    thisisoliposted 5 years ago

    @Paul the PR algorithm is not too bad, but it is only part of Google's link analysis in the SERPs algorithm.

  11. LuisEGonzalez profile image80
    LuisEGonzalezposted 5 years ago

    Has anyone asked or has heard anything from staff regarding this sudden drop? As far as my situation as of right now I have had only 8 views, when I normally would be at about 715. This has to be some sort of issue and the staff should at least try to address it and let us know, either that or they are as lost as we
    (at least I am) are......hmm

    1. PaulGoodman67 profile image89
      PaulGoodman67posted 5 years ago in reply to this

      That's what I was thinking, Luis.  Early on when there were just a handful of plungers, Paul E and Jason were active in the forums and arguing that it was individual "awkwardly written" hubs that were causing the problems.  I personally think that it's clear that the problems are deeper and wider now.

      I suspect that staff are staying quiet until the official line is worked out...  But it would be nice to get some sort of official statement, even if it's just an elaborate version of: "We're looking into it."  I can appreciate it's always difficult in these situations, however, as they won't want to give out bad info or get into arguments with upset hubbers which serve no purpose.

      I do feel a little gloomy about the future of HP, however.  I really do like it here, but I can see why many of the "HP experts" have left (even if their hubs often remain).  If Google are serious about taking down "content farms", it will be a big challenge to stop them doing so.

      1. toygurus profile image60
        toygurusposted 5 years ago in reply to this

        I  agree.There is something wrong when you are getting 700 views one day and the next day you get 8. You can't be experiencing that level of traffic then suddenly be cut off if it was merely quality. That would mean the google bot thought you had quality one day and the next day it thought the same content wasn't quality. We aren't talking about quality here. We are talking about a sudden and abrupt drop in traffic in a matter of a day.

      2. toygurus profile image60
        toygurusposted 5 years ago in reply to this

        Here' s the weird thing I have another hub account that is soaring at the moment. The quality is almost identical. So there is definately something effecting some accounts, I'm still trying to figure out what it is.

      3. Jason Menayan profile image61
        Jason Menayanposted 5 years ago in reply to this

        We are paying attention to it, but our hands are tied as to what we can do to help you. You are talking about Google changing traffic to your Hubs; we have no control over that. Most people are bouncing up, although this might not be a consolation if your traffic has tanked recently.

        The only thing we can do is to share information with each other. We do have a plan to get a deeper analysis on what comprises good quality content (in Google's eyes), but all we have now are some basic conclusions from tests that weren't particularly statistically rigorous (and which some people got angry with): don't overoptimize your content, write with a natural voice, don't keyword-stuff, publishing a bunch of doorway pages might cause problems, etc.

        On the other hand, you can look at IzzyM's example: she lost her traffic for about 3 weeks, and without any changes to her content (correct me if I'm wrong, Izzy), her traffic went back up. I think Marisa Wright's was similar.

        Which approach are you going to take if you face a traffic plunge? Neither comes with a guarantee.

        I figure the haters will begin to bash me within minutes of posting this, but, hey. smile

        1. IzzyM profile image87
          IzzyMposted 5 years ago in reply to this

          My traffic did go back up, not just back up but BEYOND back up. I was still in a state of disbelief when it collapsed again less than two days later.

          I am beginning to formulate in my own mind what may have caused the collapse, but will not know until I have made changes (yet again!!) to a load of hubs.

          Even then, if traffic recovers I will not know that I am correct because traffic is fluctuating wildly for a lot of people, not just me.

          e.g., some of us had a rise a few days ago, some of us had just changed this, that or the other, did these changes cause the upswing?

          I think not, because I didn't make those changes and I had an upswing.

          I don't blame Hubpages at all in this. This is all down to something else, something which I am sure some of us will figure out.

          I am actually sorry I asked Paul for help. Still, lesson learned, never again, but having said that, there as something in what he pointed out still jarring with me. He had a point, though I hate to admit it.

        2. Wesman Todd Shaw profile image92
          Wesman Todd Shawposted 5 years ago in reply to this

          I'm interested and at the same time. . . .blown away by your "doorway pages" link.  All I know how to do is. . . . .write text, and use hubpages. 


          Very interesting link!  I guess I'm glad that I don't know any "tricks" as I'd be dumb enough to try to use them. . . .

    2. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      Luis,it looks to me as if all your keywords have not just fallen, but disappeared! Have you looked at Google to see where your hub positions are?

      The day my traffic rose, which was on Saturday, just a few days ago, I noticed one hub had 6 visits.

      Now at that point it didn't even have 1 red up arrow, never mind 2, but I knew it had never had 6 visits in its life outwith HP new hub views.

      I searched for it on Google and found it on page 2, which surprised me, because usually I can't find my hubs.

      So then I put my best hubs into Google, and found them all at #1 or 2, and then I knew I was on a massive rise.

      They came within hours, but also within hours came a massive collapse.

      Not as low as yours, but I have one hub that gets consistent traffic, thanks to someone else who posted it to a PR5 site.

      It wasn't me, but it does tell me a Helluva lot about the value of high PR backlinks.

      1. LuisEGonzalez profile image80
        LuisEGonzalezposted 5 years ago in reply to this

        IzzyM: My keywords are still there and I checked with Mozilla SEOQuake and most of my pages are ranked 1 or 2 with some at 3. I am at a loss as to what is happening. Some of my hubs are still on page 1 or 2 with a few as the number one on the first page.....hmm

        1. IzzyM profile image87
          IzzyMposted 5 years ago in reply to this

          Have you looked at your hubs and their keywords through Scroogle.org?

          1. LuisEGonzalez profile image80
            LuisEGonzalezposted 5 years ago in reply to this

            Just did and they do show.

            1. IzzyM profile image87
              IzzyMposted 5 years ago in reply to this

              All I'm suggesting is that your keywords have collapsed. That would explain your loss of ranking and therefore traffic.

              It happened to me, overnight, three times now if we include the initial Panda shock.

              If it had happened six months before, I would not have noticed the difference.

              Although I think of myself as fairly intelligent, my first few hubs did not do well, at all.

              Then I learned a little about SEO, made some changes, and those hubs showed up on Google's front page.

              I built on this. I tweaked existing hubs, I wrote loads more. While I was never one of HP's success stories, getting views by the thousand, I was on the climb.

              Then Panda struck.

              I'm scrabbling to stay afloat as much as any other. I still thought we had a team.

              The responses of some long term hubbers show we haven't.

              I'd like to build a new one, but I suspect it can't be done. Some people, it is true, will always be envious of others.

              I can't deal with that. I honestly can't. It takes me all my time just to write hubs that pass my own personal 'quality control'. In fact, some of them don't, but I'm unsure at this point whether to target those hubs or not.
              I'm a long way from being 16.

              1. wilderness profile image94
                wildernessposted 5 years ago in reply to this

                Oh, I don't know Izzy.  Look at the response to the plunges and surges thread.  Where Marisa stepped up and put out the effort to write a hub.

                Some, both surgers and plungers, are wanting to work together and offer what help they can as well as moral support.

                Buck up, ignore those that contribute only negativity and carry on.  We (HP) did it once with the subdomains and it can be solved once more.

                Side note - like you my early hubs did not do well.  I even wrote one on intelligence (review your post) that flopped big_smile.  Unlike you I've never been able to tweak any of them into more than a few visits per week though.  Maybe just not as smart! smile

  12. thisisoli profile image64
    thisisoliposted 5 years ago

    I'm actually on the rise at the moment, seeing about 20% shy of pre-panda levels, it could drop by 90% again at any time, tick tock tick tock.

  13. Wesman Todd Shaw profile image92
    Wesman Todd Shawposted 5 years ago

    I can't help but to have also noticed this . . . .:=\

    Hats off for IZZY!!!!

    But I have to admit that I was thinking about the poster that I, at least, am thinking of. . . .

  14. Richieb799 profile image64
    Richieb799posted 5 years ago

    I agree, someone with a big collection of hubs like Izzy's should be looked at as a great sub domain by Google, I'm sure it will stand tall when all of the turbulance settles down. I've found alot of good web 2.0 sites which I used for backlink at the moment

  15. Marye Audet profile image91
    Marye Audetposted 5 years ago

    I get that. I am not sayng anyone is a crap writer... I don't have time to read many articles that I am not being paid to edit, quite honestly. What I was trying to say, and probably failing miserably, was that it takes time to get to the point where you have authority with Google, whether it is articles, blogs, or your name as an author but it does happen and it is worth the effort it takes to network, market yourself, and get your name out there. Don't be dependent on the authority of your blog, or website, or the client you are working for - create authority with your name, so that Google sees you as an "expert" on something. Then when the crawler thing catches your name on an article it takes note of it. That's why it is best to use either your real name or use the same nom de plume always.

    Not many people became successful overnight. And IzzyM, yeah, I do seem to catch a lot of flack on the forums and in the emails I get fromother hubbers about being stuck up... Which, if you knew me, would keep you in gales of laughter for a long, long time.

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      While I thought the rest of your comment was just lovely, this bit stuck in my throat and I had to reply.

      I have seen a few long term hubbers report something similar. They won't reply or write in the forums any more because of flack received ( I was looking around for another word but couldn't find better).

      I am so sorry. It wasn't me and would never be me, because that is not how I approach things.

      I never saw you as stuck up, but if I did, I would tell you.

      Nicely, of course smile

    2. Smart Rookie profile image74
      Smart Rookieposted 5 years ago in reply to this

      I'm really glad you took the time to expound on your previous statement. Not knowing you, I read a bit of condescension into it - not that I was going to say anything; I tend to keep quiet and just read on. Of course, being on the losing end of this deal has made me more sensitive than I would be if I was sitting pretty.

      What you've said about a name and a visible presence contributing to authority makes a lot of sense. It will definitely change how I proceed. I've got a lot of tearing down and rebuilding to do.

      1. IzzyM profile image87
        IzzyMposted 5 years ago in reply to this

        You can't tear down and rebuild unless you know what is wrong.

        There are some points to be made here, but sit a while and think about them first.

    3. Wesman Todd Shaw profile image92
      Wesman Todd Shawposted 5 years ago in reply to this

      Well I offer an apology then in light of the more defined statement.  Sorry if I made you feel a target.

      1. Marye Audet profile image91
        Marye Audetposted 5 years ago in reply to this

        Sorry I snapped. I am usually more patient. But today Aline I have had to write to the editor of a largish newspaper and tell them that their author wrote an interview with me without ever contacting for an actual interview... They just took info from a variety of my posts and articles... A ex that doesn't believe on child support.. Two kids in college.. A broken gas line that will cost 5 k to fix...a 1600.00 gas bill because of that line... 25 websites thSt copied my content, making over 100 dmca complaint so far this month... a stalker that now calls me on my phone... And an ex client that is replacing my name on articles I wrote with someone else's...there's more but that is all I feel like writing. smile Sometimes it takes me a lot of effort to stay positive so when I come to the for and try to be encouraging it hurts to have people think I am being boastful or stuck up. Sometes I am just too tired to step back and count to 10. I am sorry for my unkindness.

        1. IzzyM profile image87
          IzzyMposted 5 years ago in reply to this

          Awwww (((((hugs, babe ))))))

          Tomorrow is another day smile

        2. Marye Audet profile image91
          Marye Audetposted 5 years ago in reply to this

          Alone ... today ALONE... stupid phone spell check...

  16. Wesman Todd Shaw profile image92
    Wesman Todd Shawposted 5 years ago

    I think the "Authority" comment came across as rather superior.  Yes, I of course know all about Google assigning author's authority.  I think a first time Hubpages publisher out of nowhere, however, is as likely as the, ahem, "elite" of this site to put out something great.

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      I was half-way through writing a reply to someone else who commented on this thread, someone who had only been here 11 months, when their comment 'disappeared'.

      I'm not into Alice and Wonderland and 'down the rabbit hole' stuff. Speak to me a plain as daylight.

      Got something to say? Say it, and tomorrow be damned. (That can be taken two ways at least).

  17. Will Apse profile image89
    Will Apseposted 5 years ago

    It is curious, though, that Mayre attracts these attacks. It is good to have the input of successful writers in these forums. And it shouldn't be a prerequisite to have skin as thick as an elephants.

  18. Maralexa profile image88
    Maralexaposted 5 years ago

    Is there any advice you good people could give to new hubbers = those who have been writing and trying to monetise their 'sites' since the sub-domain switch?  I can only guess at the work you must need to do to reposition your 'sites' and keep the benefits you had in the past.  I truly wish you all real success.

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      Maralexa, my advice would be to write your hub as naturally as possible in Word or whatever software you use. Then, like a lot of us, spend a lot of time and careful consideration over the title, and thus the url of that hub.

      Try to think of words that someone might type into a search engine, on your subject matter.

      Be as accurate as you can. If your hub is about how to do your Geography homework on the internet with minimal work, then title it as such, but limit your title to 70 characters. It is 70?

      Assuming, of course, people are going to type into search 'How to do Geography homework'.

  19. Maralexa profile image88
    Maralexaposted 5 years ago

    Thank you for taking the time, Izzy.  Your comments are truly appreciated.

  20. Will Apse profile image89
    Will Apseposted 5 years ago

    This video seems to suggest Google is using domain blocking data in conjunction with Panda. It is pushed out in a hurried aside.

    http://www.dailymotion.com/video/xkx0th … panda_tech

  21. Silver Rose profile image77
    Silver Roseposted 5 years ago

    People need to calm down. These reports of alternate plunges and surges reminded me of something that happened in June just before Panda 2.2 was released.

    I went and found the thread on the Warrior forum:

    http://www.warriorforum.com/adsense-ppc … ely-3.html

    Google was A/B testing. For a summary it went something like this:

    6th June

    Group A (site a,b,c) Back to Page 1,2
    Group B (site x,y,z) Gone

    10th June

    Group A (site a,b,c) Gone
    Group B (site x,y,z) Back to Page 1,2

    14th June

    Group A (site a,b,c) Back to Page 1,2
    Group B (site x,y,z) Gone

    17th June

    Group A (site a,b,c) Gone
    Group B (site x,y,z) Back to Page 1,2

    Panda 2.2 was released on 17th June and Group B "won".

    The reports of surges and plunges we are seeing starting on 23th August sound very much like they are now a/b testing the new subdomains.

    What we need to do is isolate the dates. I haven't had my plunge yet, but mine is a small account. Does anyone want to chip in with dates?

    1. Susana S profile image91
      Susana Sposted 5 years ago in reply to this

      The theory that Google is doing some kind of testing makes sense to me after spending hours looking at my analytics stats and pretty much coming to the same conclusion. Since some reports from those clever enough to understand google patents indicate that Panda is a learning algorithm, it would make even more sense that this is what is happening.

      My opinion at the moment is that Google is learning how to match search queries to serp results. The reason I think that is because when I look at the keywords bringing extra traffic to my hubs since my latest surge, the vast majority are longtails. Here's an example:

      One hub that used to get the majority of its traffic for a few (optimised for) phrases now has far more traffic coming from a variety of longtails each with 1 or 2 searches per day. This has taken that hubs traffic from 10 hits per day to over a 100. Nearly all of the extra traffic across my account is coming from these low volume longtails. It's the same with my other hubs. One that used to get about 1500 per day is now getting 3,500 and again all the extra traffic is from longtail phrases.

      I expect Google will be looking at which phrases best match the page by looking at the time on page and bounce rates (plus other things I'm sure).

      So back to analytics - the majority of these long tails that are delivering this boost in traffic are showing zero time on page, so I imagine that it won't be long before goog decides (learns) not to rank my page for those phrases and traffic will go down again.

      My dates are:

      10th August

      6th September

      1. Silver Rose profile image77
        Silver Roseposted 5 years ago in reply to this

        Remember that the bounce rate in your analytics is different from the bounce G is looking at. Your analytics only reflect that the visitor did not go to another page on your site. However, they could have clicked through to amazon or clicked on an ad.

        All G is interested in is bounce back into the SERPs.

        My guess is that they need a standard amount of SERPs impressions (a search result where your site is listed) to make a good quality assessment - say 250,000 impressions. Then they look at how many people click through from the SERPs to your site, how many people backspace back into the SERPs, and if they backspace, how long the interval is before they do so (i.e. did they read or did they decide within a few seconds that the page is not what they wanted).

        So far, I believe the different experiences people are having are just down to the size of their account and the number of keywords they are targeting, and the type of keywords they are targeting. Some are going for large vol low CPC words, and others are going for low vol high CPC words. That in itself will affect how quickly G gathers it's data.

        I haven't plunged yet - but I don't think this is to do with my uber stellar writing ability! I think it's because they haven't got their full sample size from me yet as my account is small. People with a large amount of hubs with large traffic are going through the process faster.

        I guess we'll know in a week or so who has passed the test and who has lost.

        1. Susana S profile image91
          Susana Sposted 5 years ago in reply to this

          I understand bounce smile When my analytics are showing zero time on page and 100% bounce for certain phrases over a fair number of page views, I'm pretty sure my page will soon stop being listed on page 1 for those phrases. Possibly even if they are bouncing due to clicking an ad - I don't think the search quality and adsense teams work together very much.

          I think your thoughts about the time taken to get enough data depending on number of impressions makes sense in terms of how quickly or slowly accounts might plunge and surge. Not sure it is actually playing out that way though hmm Some accounts with less than 1000 views per day had a surge for just a few days, while others with a lot more views per day have had high traffic lasting weeks (and in many cases has continued to stay high up to this point).

          That could *possibly* be explained by the percentage of click throughs from the serps for a particular website (in this case our subdomains) - maybe if a site doesn't reach a critical threshold of click throughs it gets dumped further back in the serps? That would have nothing to do with quality of writing only with how appealing the title and description are to the searcher.

          So, so, so many variables - it really is mind boggling (to me anyway) tongue

          I would love to know if anyone else who is experiencing increased traffic is seeing the same thing as me with the longtails?

          1. wilderness profile image94
            wildernessposted 5 years ago in reply to this

            I looked at a few of my better performing hubs and sure enough, there are quite a few new long tails there.

            Nearly all were good matches, though.  The only hub that had more than 1 or 2 that were probably wrong was a hub on page 2 - the others were all on page 1.

            It makes very good sense to me that google is testing the hubs but it leaves the surgers that don't plunge kind of high and dry.  Either we haven't been tested or we have been tested and every time pass with flying colors.  It's hard to believe either option.

            While I think this idea has a lot of merit, I'm still looking at those massive numbers of new external backlinks.  Backlinks that appear overnight, change daily and have little to do with relevance (I found a link to my hub on wiring a 3 way light switch on a hub by Prettydarkhorse about giving birth the natural way with Lamaze!?).

            One piece of evidence that I haven't seen at all is the other SEs.  Does anyone out there get a good bit of traffic from Bing or Yahoo?  If so, are those numbers cycling too, did they jump with the subdomain (all all those new external links) or was there no change at all?  I get a few visits from each, but nowhere near enough to provide reliable data.

            If all three are cycling it could indicate that backlinks are the big problem; if not either that the other two don't crawl often or that the Panda is testing us.

            1. Will Apse profile image89
              Will Apseposted 5 years ago in reply to this

              I would love to feel that I had been tested and found pure and wholesome.

              It would be a first.

              1. wilderness profile image94
                wildernessposted 5 years ago in reply to this

                Me too!  Unfortunately, the feeling that I have merely dodged the arrows to date is much, much stronger.  Sooner or later I will be hit.

            2. Susana S profile image91
              Susana Sposted 5 years ago in reply to this

              Thought I'd take a look at Bing and Yahoo traffic on my best performing hub and compared 3 days (tues, weds, thurs) in june, july, august and september. The dates I chose covered the traffic extremes and pre and post subdomain.

              Traffic from both Bing and Yahoo is between 30 - 40 % less since the subdomain switch on that hub, but stable at that level. So no cycling, just a reduction.

        2. Aficionada profile image93
          Aficionadaposted 5 years ago in reply to this



          I thought wilderness's theory about external links had a lot of merit, until Silver Rose posted such a sensible and reasonable probable explanation, which (I think!) seems to cover all the variables that have been reported.

          My experience is very similar to SR's.  Our accounts are about the same size.  I have only seen a surge so far - no real plunge yet, although there have been some minor dips; but my numbers, small though they are, just keep going up - currently about 50% higher than when Panda 1 struck.  I'm holding my breath.  I guess I'll know when G is ready to test me. sad

          I checked my Yahoo! and Bing statistics (in the HP account stats), comparing them to Google.com views over the life of the account and comparing their performance in the last week and the last 24 hours.  I was surprised that the ratios have remained fairly stable - except for a strong dip in Bing over the past week.  I don't know how meaningful this is with my small numbers (but I'm very glad to get any Yahoo! and Bing traffic).

          I donwannaplunge!

    2. Marisa Wright profile image93
      Marisa Wrightposted 5 years ago in reply to this

      This is the first decent theory I've seen that makes any sense at all! 

      So what you're saying is, we may as well all settle down and wait for the next time Panda is run - then we'll see what decision Google has come to.  Right?

      1. Silver Rose profile image77
        Silver Roseposted 5 years ago in reply to this

        Right.

        They are gathering data, and they have to remove you from the SERPs for a period to assess whether it is better without you or worse without you.

        People need to sit tight and wait it out - fiddling with stuff now might mean that you are messing with a winning formula, and when they re-include you for iteration 2, they find you arn't as good as indicated in iteration 1!

        Basically wait until the final Panda is run and only if it's clear that you have definitely "lost" should you then attempt to correct stuff.

        1. Sally's Trove profile image97
          Sally's Troveposted 5 years ago in reply to this

          Agree...this explanation makes more sense than anything else I've heard since Panda hit. However, I read somewhere (sorry, can't remember where!) that Panda will continue to be run from time to time; in other words, there will not be a final Panda, only a next one and a next and a next...I guess, until Google runs out of eucalyptus.

          1. Silver Rose profile image77
            Silver Roseposted 5 years ago in reply to this

            I should have written "wait until it has been announced that Panda 2.5 has run". Because you are right, they will have more panda's in the future. But I don't think the testing will be as savage - we're experiencing this because they have no data on the sub-domains at all, they are "new" to the machine.

            It's very tempting to think that your writing is bad or you've done something wrong, when you've been removed from the SERPs during testing. But the machine is completely neutral, it's just trying to get a snapshot of user satisfaction without you, to compare to how users react with you included. It's painful and scary but it's the only objective way of assessing things.

            So don't take it personally if you've disappeared temporarily and people should also refrain from making value judgements of those who have gone from the SERPs for a bit - it's just way too early to say what Panda will decide till the test is complete.

            1. Lisa HW profile image80
              Lisa HWposted 5 years ago in reply to this

              As others have said, this, to me, makes a lot of sense and makes any of the "weirdness of it all" seem a whole lot less mysterious. I've pretty much been more been concerned by the "weirdness" (and varied cases of it) than by the prospect of being "stamped a loser" by Google.

            2. PaulGoodman67 profile image89
              PaulGoodman67posted 5 years ago in reply to this

              I hope that you're right.  I certainly don't plan to change much, myself.  I am reasonably happy with how my writing does in non-HubPages sites, so I don't really see my writing as a major issue. 

              Still, I do worry that some people could fall through the cracks of a New World Order.

        2. Susana S profile image91
          Susana Sposted 5 years ago in reply to this

          Yes, that makes good sense. But for what keywords? I think that's the other side of this. What I'm seeing while my traffic is surging is that my pages are getting thrown up on page 1 for anything and everything that could possibly related to the page - hence the large volume of longtails - most of which are only loosely related to the page and most of which are getting zero time on page and 100% bounce.

          Fortunately, the main keywords I have targeted get a long time on page, so hopefully when Panda runs again my pages will remain high in the serps for those phrases. The extra traffic is great for HPad earnings, but it's not the right traffic in terms of search query/page match. I expect to lose all of that extra traffic very soon.

          1. Silver Rose profile image77
            Silver Roseposted 5 years ago in reply to this

            Your page doesn't have to be "perfect" for a particular long-tail - it just has to be better than all the other pages served up for that keyword. The machine is not sure, so it is running a test and giving you the chance to prove yourself and perhaps rank for keywords that the old algo wouldn't have picked you for.

            You are right, for some it will drop you, but for others it might go "ah ha, that page is just what we needed". Who knows?

            When our hearts have stopped racing from all this stop-go excitement, it might be that panda and sub-domains are the best things to have happened to us.

            1. Susana S profile image91
              Susana Sposted 5 years ago in reply to this

              No not perfect - but better than 0.00 on page and 100% bounce I would have thought, lol. Will have to wait and see if any of those getting a bit more time on page will stick....or not.

              Whatever the causes, reasons, whys and wherefores I am truly fed up with the highs and lows. I can't imagine at this point ever being grateful for Panda or subdomains.

              1. gracenotes profile image93
                gracenotesposted 5 years ago in reply to this

                I've never really understood the 0:00 time on page.  How can it be zero seconds?  I see it a lot in GA.

                I did some research, though, and the average amount of time searchers spend on a page is about 10 seconds (I got this statistic from the book THE SHALLOWS: WHAT THE INTERNET IS DOING TO OUR BRAINS). Perhaps Google looks at that as a quality marker.

                1. thisisoli profile image64
                  thisisoliposted 5 years ago in reply to this

                  Because of page load time GA is not 100% accurate, stats should be a little better with asynchronous code. 

                  Generally 0 seconds means they bounced pretty much immediately though.

                2. Susana S profile image91
                  Susana Sposted 5 years ago in reply to this

                  What Oli said.

                  No one really knows if time on page is a quality marker, but many are theorising that it has become more important following the panda update. Just found a great article about time on page here: http://www.searchenginejournal.com/tick … ent/21439/

                  It kinda blows my theory out of the window though hmm

                  "...if someone reaches a certain page and exits from that page (or bounces), then Time on Page is 0:00.  No, that’s not a typo.  Time on Page is 0:00 whether they spent 10 seconds on the page or 10 minutes."

                  1. thisisoli profile image64
                    thisisoliposted 5 years ago in reply to this

                    Great article, thanks for linking!

        3. Wesman Todd Shaw profile image92
          Wesman Todd Shawposted 5 years ago in reply to this

          I like your thinking! :-)

  22. Silver Rose profile image77
    Silver Roseposted 5 years ago

    From Marisa's hub

    http://marisawright.hubpages.com/hub/Hu … ges-survey

    The dates seem to be

    August 10th

    August 23rd

    August 29th

    Sept 7th

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      I would change that last date to September 6th. Some may not have noticed till the 7th, but the fell came on September 6th.

      1. PaulGoodman67 profile image89
        PaulGoodman67posted 5 years ago in reply to this

        My big dramatic fall came on Sat 3rd Sept, I double-checked to make sure I was right.

        1. PaulGoodman67 profile image89
          PaulGoodman67posted 5 years ago in reply to this

          The 23rd August was the last significant date for me before that for me, as my traffic surged then.  I find Silver Rose's idea that people are passing some sort of threshold before they plunge is maybe more feasible than that there are indivdual algo changes happending every few days. 

          It really isn't worth me doing much on HP until something changes.  My Google.com is sub 5 views. at the mo

        2. IzzyM profile image87
          IzzyMposted 5 years ago in reply to this

          On that same date I had a surge. Shortest surge in history maybe as it fell a day and half later.

  23. Anna Marie Bowman profile image90
    Anna Marie Bowmanposted 5 years ago

    All this bickering...ugh...I hate it.  Everyone pointing fingers at everyone else, jumping down other's throats.  I understand this is frustrating.  I still don't understand a lot of the SEO stuff, the various other algorithms, etc.  I do know that when I started, writers like Mayre were an inspiration to me, a motivator for me, and a wealth of information on a lot that I didn't understand.  Over my three years here, I have learned a lot, and still have a lot to learn.  I write because I love to.  I wish I were able to write and make enough money writing, like she does.  She is still an inspiration to me, someone to look up to, an indicator of the goals I still have to achieve.  Over the years I have seen many Hubbers come and go.  A lot of the ones who have been long gone, were prone to attack the experienced Hubbers who should have been a source of inspiration and advice, instead of a face to place envy and resentment.  Could I look at that opening page, seeing the same faces there every day, and feel resentment that my face isn't there?  Of course.  Does that do anything to improve my writing?  Not even a little bit.  What improves my writing is my drive to get my face on that page. 

    I have done a lot of Hub Hopping lately, it is an easy way to see what new writing is popping up, what new Hubbers are out there, and gives me something to do when I am experiencing a little bit of writer's block.  I have found a lot of sub-par Hubs.  And I mean a lot!  A Hub of one small paragraph is not a quality Hub.  I know that there are a lot of great new writers out there, and I have found many through my hopping.  I know the point of all these changes was to highlight talented writers while weeding out the sub-par writing on the site, and therefore improving the reputation of HubPages and the writers here.  Has it negatively affected a lot of talented writers, inadvertently? It seems that it has.  With everything new, it takes a little time to work out the kinks...things like this have happened before, and they will happen again.  Please, to all the great writers out there, just be patient...ride out the rough waves, and calmer seas will follow.

    1. David 470 profile image87
      David 470posted 5 years ago in reply to this

      Nice speech lol

    2. Marye Audet profile image91
      Marye Audetposted 5 years ago in reply to this

      Thank you, so very much. smile

  24. PaulGoodman67 profile image89
    PaulGoodman67posted 5 years ago

    As the person responsible for starting this thread, thanks to everyone who's contributed so far!

    Silver Rose's theory that Google is testing is the theory that I personally find most appealing so far, to be honest.  As a plunger, it seems striking to me that my google.com traffic is pretty much zero.  We're not being given lower rankings, we're being pretty much taken out of the equation altogether in many cases it seems (hopefully it will be very temporary, but who knows?)

    I really do think that there is a "technical" reason in this case for all the plunging and surging, rather than it being because someone writes better, or someone writes worse, or they use lots of SEO, or they don't use SEO, etc.

    Still, everyone is entitled to their own opinions, of course.

  25. Will Apse profile image89
    Will Apseposted 5 years ago

    If this is testing, why are the top Hubbers (traffic-wise) more or less exempt?

    According to Paul Edmondson, only 2 out of the top 50 have seen any volatility after the huge rises in volume recently.

    1. wilderness profile image94
      wildernessposted 5 years ago in reply to this

      Is it at least partially a function of the size of the subdomain?  We've seen some large numbers here, like Izzy and a few others, but there are some that are far, far larger and, presumably, get a lot more traffic.

      I can only recall seeing one huge hubber with 1700 hubs and that one report had no cycling - just a surge.

    2. toygurus profile image60
      toygurusposted 5 years ago in reply to this

      Not sure if this has been posted but google recently uploaded a video as to how they tweaks to their algorithm. They don't give the same results to everybody. They only give the results to a small sample size when making a tweak to their algorithm. You can find it here http://www.youtube.com/watch?v=J5RZOU6vK4Q

    3. Fiction Teller profile image61
      Fiction Tellerposted 5 years ago in reply to this

      Two possibilities I can see:

      1) Google doesn't do everything at the same time, but rolls things out in waves.

      2) Google's sampling rather than sweeping, to get a sense of what HubPages now is as a whole.  Polls work that way. Well-designed ones take a small portion of the population and make accurate predictions.

      We still don't really know to what extent our subdomains are treated as separate websites to Google. Not every site on the web goes through Panda. Many personal blogs and websites? Not even run through Panda.

      Not every query is necessarily affected by Panda. ("Cheap laptops" will probably be treated differently from "who was the first president of the united states".)

      Not every CLASS of queries is necessarily affected by Panda. (navigational searches vs. informational searches, searches in different topic areas, etc.)

      I'm also liking Silver Rose's theory of A/B testing for some algo change, maybe Panda, maybe something else. Google's moving toward using user metrics in unprecedented ways. For this, they need data. How much varies from query to query.

      1. David 470 profile image87
        David 470posted 5 years ago in reply to this

        Why would Google not want to treat our subdomains as separate webpages?

        I sure hope they don't look down negatively upon these subdomains, as I thought it was a great idea because spammers won't screw good writers over.

        1. Fiction Teller profile image61
          Fiction Tellerposted 5 years ago in reply to this

          Not saying that. Just that Google probably has a sliding scale in how it views the relationship of subdomains to the master site. I remember reading some time ago that Google treats subdomains as separate sites IF they really seem to be different sites - different looks, different layout, etc. Too many similarities to the main site, and they treat them as the same. There could be levels of relatedness in between those two extremes.

          Wordpress.com and Blogger are different from HubPages.  Unlike on Blogger, we don't have total control over stuff like the links, arrangement of content, and appearance. We have the use of a limited number of modules for our content, and some types of content, formatting-wise, are not possible. Google might or might not take into account the fact that we don't have full developer-level control of our subdomains and they all bear a remarkable similarity to each other.

          1. PaulGoodman67 profile image89
            PaulGoodman67posted 5 years ago in reply to this

            HubPages structure is unique, being split into subject areas and then subdomains.

      2. Susana S profile image91
        Susana Sposted 5 years ago in reply to this

        Completely agree with this.

    4. Marisa Wright profile image93
      Marisa Wrightposted 5 years ago in reply to this

      The last time Paul made a comment was well before some of us started to experience the plunge/surge phenomenon.  He did make another comment on my Hub but I'm not clear whether they had done some new research, or whether he was basing the comment on the work they did when Izzy was in the plunge phase.

  26. sharewhatuknow profile image78
    sharewhatuknowposted 5 years ago

    Whoa, there is so much information here and so many replies to read that it will take 2 days.  Not really, but this topic is a powerhouse of information.

  27. PaulGoodman67 profile image89
    PaulGoodman67posted 5 years ago

    One argument against the "Google testing" theory though, might be, or might become the timescale.  Izzy and Randy have been plunged since the 23rd August - the longer it goes on, the less likely it would seem to be just being temporary testing.

    1. IzzyM profile image87
      IzzyMposted 5 years ago in reply to this

      No it was the 10th of August I plunged, and I did have a brief respite last weekend when it shot up to record levels before plunging again.
      I'm keeping a very close eye on stats for signs of a recovery and today I am seeing keywords falling further in the SERPS.
      I am now in the process of re-writing all my summaries as many of them were done in a hurry without much thought or care.

      1. PaulGoodman67 profile image89
        PaulGoodman67posted 5 years ago in reply to this

        Ah, sorry for getting date wrong. 

        I opened a thread asking if anyone had recovered from plunge.  I am not sure that anyone has in any big way.  If google introduces another update this month and there is no recovery, or more people plunge, then it will be time to start getting very worried.  I am not sure where HP would go next in that situation.  I do wonder if in private, Google would consider taking down HP as a major "content farm" scalp.  I hope that I am completely wrong on that. 

        The summaries are a sideshow, aren't they?  Or are you just filling up time?

        1. IzzyM profile image87
          IzzyMposted 5 years ago in reply to this

          You are joking, of course. Summaries are extremely important!

          I found that out as a newbie when I was first learning (about six months in). At that time to write the summary, all one had to do was make sure the first sentence or two in the hub contained your keywords.

          When I changed mine, I saw real traffic for the first time.

          Then when they changed over to separate summaries, which was months ago, long before they put the A sign on our hub stats, I had to re-do all my summaries and to be honest I just fired through them without much thought.

          Now I am re-doing them, with thought, so we'll see what happens, if anything.

          I'll have 400 odd more views showing today if nothing else! (my own)

          1. PaulGoodman67 profile image89
            PaulGoodman67posted 5 years ago in reply to this

            No, you misunderstand.  I meant that there was no connection between the summary and the plunge, in my opinion.  Some people have blamed HubPages' attempts to persuade people to write summaries with the plunge.  (Google is very much the problem here, methinks.)

            The summaries will only affect google traffic, methinks, and my only traffic is links.  I doubt very much that rewriting summaries will have any effect, in this climate but good luck.  I am not working on HP until the plunge period is over and focusing on my other sites instead, as they seem unaffected by Google.

            I want to spend less time on the HP forums too, but am failing at that so far!  ;-)

            1. profile image0
              Multimanposted 5 years ago in reply to this

              I got a message from team hubpages sayingthat these up and down will continue for a couple more weeks.

            2. toygurus profile image60
              toygurusposted 5 years ago in reply to this

              Seems like a good idea Paul. I don't think being on the forums or writing new hubs is going to accomplish much. We can only wait and see. If traffic returns, it returns. Hopefully Team Hubpages is right and it is only a temporary blip for you.

              1. PaulGoodman67 profile image89
                PaulGoodman67posted 5 years ago in reply to this

                :-)

            3. toygurus profile image60
              toygurusposted 5 years ago in reply to this

              I'm no expert on search engine rankings, but I have seen this pattern before on my own websites. It seems like google is treating each subdomain as an individual brand new websites. Initially when you create a website you get a lot of traffic. Then google drops it in the rankings and reevaluates it. This is called the honeymoon period. If your website has quality content it should return in a few weeks or months. I am hoping this is what is happening with hubpages.

              1. PaulGoodman67 profile image89
                PaulGoodman67posted 5 years ago in reply to this

                I know what you mean.  But there doesn't seem to be a clear pattern that fits that though, unfortunately.  Mine and your traffic might fit that idea (although the timescales seem more than a little off), but there are others who've had wild swings up and down.  It doesn't fit with the pattern that I've seen when setting up websites either.

                What is weird is that our hubs haven't just dropped down the rankings, they are pretty much out of the rankings altogether.  Which is why google traffic drops to virtually zero.

                Anyway, nothing much can be done by us.  I will get over the shock, I'm sure, and things will either get better or they won't! smile

                1. Greekgeek profile image98
                  Greekgeekposted 5 years ago in reply to this

                  One pattern I do see is that Hubpages shows a modest but measurable uptick traffic on the Quantcast graphs after Google pushed the "Panda button" around Jul 25.  (Google "Panda 2.3" for discussion of this across the SEO punditsphere.)

                  A good sign. I have a hunch Hubpages subdomains may have helped. Matt Cutts made noncomittal noises about them, but it stands to reason that if Panda is assessing domains to see the spam/quality ratio of each domain, it may be doing the same, to a smaller extent, with each subdomain, especially where the rel=author metatag on each Hub makes it clear that each subdomain is a different author.

                  So then we just had to wait until the Panda button got pushed again so that Google recalculated the "Panda Factor" for subdomains on HP. If not everyone had switched to subdomains yet on Jul 25, we may see another small jump the next time Panda gets rerun.

              2. Marisa Wright profile image93
                Marisa Wrightposted 5 years ago in reply to this

                Exactly, because that's what they are.  However the pattern isn't quite the same, with some people experiencing very steep climbs and plunges, and with more than one peak and trough.

          2. Marisa Wright profile image93
            Marisa Wrightposted 5 years ago in reply to this

            I agree they're important but most of my Hubs don't have them. 

            I didn't realize how many I hadn't done, until the "A"'s appeared!  I haven't gotten around to fixing them but despite that, my traffic has taken an upswing again - so I'm not sure summaries have a lot to do with anything.  But it's quite possible I might be doing even better if I fixed them!

          3. thisisoli profile image64
            thisisoliposted 5 years ago in reply to this

            Meta Description is not used by Google in terms of SEO, what it is useful for is including click through rate of your Google listing.

            1. janderson99 profile image87
              janderson99posted 5 years ago in reply to this

              It is used for the summary displayed on the search engine result and is what the human uses to decide whether to click on your link - so its is very very important. Otherwise Google extracts random stuff from your page to display.

  28. Randy Godwin profile image93
    Randy Godwinposted 5 years ago

    And some--myself included--experiencing nothing but a plunge.  smile

    1. PaulGoodman67 profile image89
      PaulGoodman67posted 5 years ago in reply to this

      Well, my traffic is rising at the mo, it might well just be a flash in the pan, but hey, I get to experience the red trangles!  wink

      The big hope is that the next Google algo will sort out all the turbulence and restore some sanity (esp for the 'awkward writers')!  smile

      1. IzzyM profile image87
        IzzyMposted 5 years ago in reply to this

        Good luck Paul, hope this is your turnaround.

        I see Marisa's traffic rose yesterday. Fingers crossed for her.

        My turn next hopefully. It was a week ago tomorrow I got the spike, maybe it will come back.

        1. PaulGoodman67 profile image89
          PaulGoodman67posted 5 years ago in reply to this

          I am cautious about the rise, but it's better than bumping along the bottom.  Some people in the forums, esp those that haven't crashed, don't see why people make a fuss, but the difference in earnings between having big traffic and no traffic is proportionally huge.  I am no big earner, but a few dollars here and there can make a difference to daily life.

          Spikes are no good in the long run.  I find the theory that Google are testing and things will stabilize with the next algo change (panda 2.3?) later this month appealing.  It's speculation, of course.  But  I hope all the good writers, including you and Randy, benefit.

      2. toygurus profile image60
        toygurusposted 5 years ago in reply to this

        This is good news Paul. Mine too.

  29. Aficionada profile image93
    Aficionadaposted 5 years ago

    But in regards to time on page (per the linked article):  shouldn't the brainiacs at Google be able to figure out the limitations of measured time-on-page vs. actual time-on-page?  It seems that with that limitation, time-on-page (as measured) would have lesser significance - or at least it ought to (IMO).

    1. Susana S profile image91
      Susana Sposted 5 years ago in reply to this

      Only google knows how much significance (if any) they give that metric (if we only knew, we'd have a better idea what to work on!). I agree that it shouldn't be a huge factor since it's wholly unreliable as a measure of quality, but whether it is used as a trust/quality/ranking factor, who can say?

      The other question is where they would get their data from? Chrome users? Google toolbar? Direct from search (for back clicks)? Analytics? All of these are limited.

      The more I read the less influence I think that time on page and bounce rates deserve to have.

      1. Will Apse profile image89
        Will Apseposted 5 years ago in reply to this

        Somehow, Google does work out what pages readers like. PR and links in are only ever part of the story when you look at those pages that get to the top of the SERP's. But how they do it without feeding in user metrics is a mystery.

 
working