Thoughts on traffic since Panda 3.3

1. 74

   Max Daltonposted 12 years ago

   It looks like my traffic has bounced back to where it should be. I also cranked out a lot of articles (for me, at least) after the new algorithm change as my way of dealing with it, but it looks like in spite of the incoming traffic to those new hubs, I'm still up. I was just curious what other people are seeing. I know the Panda hits affect us all differently.

   replyreport
   1. 79

      Bendo13posted 12 years agoin reply to this

      My traffic's down by about 25%-ish... but still steady.
      
      This weekend had a big dip but it's coming back up... I don't think people with hangovers like to do much reading

      replyreport
2. 97

   PaulGoodman67posted 12 years ago

   My HP traffic is pretty stable but down more than 15% on what it was, overall.  Individual hubs hit differently, however, and some actually seem to have benefited - but mostly it's not good (though not disastrous).
   
   All the "HP Success Stories" seem to be going down at the moment and you can see there was a hit on Quantcast if you look closely, although it's not as crude as the falling off the edge of a cliff graph that came with the first Panda.
   
   HP at Quancast

   replyreport
3. 93

   Dolores Monetposted 12 years ago

   I had a drop at the beginning but not like the 1st panda. Bounced back up within days. Down a tad today, but my stuff usually gets more traffic mid week.

   replyreport
4. 87

   Stacie Lposted 12 years ago

   well my traffic is still down.It usually bounces back after a few weeks.

   replyreport
5. 91

   JKennyposted 12 years ago

   My Google traffic has actually experienced a sustained increase, but then again, I'm relatively new to this game, so I'm expecting it to fluctuate.

   replyreport
6. 61

   iphone5newsposted 12 years ago

   my sites got pounded by 3.3 or I got a manual review - either way google slapped me hard.

   replyreport
7. 60

   Randy Godwinposted 12 years ago

   I was already in the sand box.  As my grandma used to say "you cain't git blood from a turnip"!
   
   
   
                                          

   replyreport
8. 62

   MakinBaconposted 12 years ago

   Another issue is how each hub performs. For example, I have some hubs that have averaged over 1,000 visits a day. If you have a couple of those get hit and the rest stay the same, you could lose up to 20 percent or more of your daily traffic, even though the rest of the hubs are performing well.
   
   I have had some of my best ones take a hit, so have experienced that across all of my profiles.
   
   In the past when I've taken a hit from Panda, it has usually totally clobbered the hub. This time around they've taken big hits, but are still generating decent traffic, making one think there may be more than one variable in the mix.
   
   The thing that keeps me suspicious is all of my best articles have been getting slammed, even though they cover completely different topics, suggesting something that appears to be targeting them.

   replyreport
   1. 78

      Greekgeekposted 12 years agoin reply to this

      Panda is domain-wide, so if you see across-the-board drops or increases, that's probably the culprit.
      
      Hubpages is gambling that dividing our content into subdomains invites Panda to treat each subdomain separately, so we won't be penalized for junk posted by other Hubbers.
      
      I'm still waiting for some kind of statistical study proving this. It makes sense, but I'd love to see the numbers.
      
      What's interesting is that Hubpages shows a traffic recovery at the end of last August, but it doesn't match up to the Panda updates for last summer: July 23, Aug 12, Sept 28. I can't figure out why Hubpages' traffic did a major saving throw around August 23.
      
      Also, the curve in the last 3 months is such that I can't tell if the wobbles are from Panda 3.3 or just seasonal variations.
      
      Little did I know that I really needed to major in stats in college. I find it all so fascinating, but I'm never 100% sure what's statistically significant. Nevertheless, I came back to Hubpages because of Panda: I was curious to see how it played out.

      replyreport
      1. 97

         PaulGoodman67posted 12 years agoin reply to this

         Like Will (below), I think that it's fairly well established that the subdomain shift brought HubPages out of the shadows of Panda and that the surge in late summer were directly related.  Paul E was talking at SXSW earlier this week on this topic, I believe, so there may be a video of what he had to say on the matter?  The success of subdomains in the longterm remains to be seen, but if HP had done nothing it would've died on its feet.  They couldn't manually check so many hubs for content.
         
         Panda 3.3 was introduced in early March so anything before that is not caused by it, although it's a bit of an academic distinction as Google could have made unannounced tweaks to the algo beforehand.
         
         One thing that has always struck me is that when Google say that they want to sort out the search results, I suspect they just mean that they want to sort out the first one or two pages of results - they probably don't care much what happens on page 7 or 17, as the public don't really notice those ones so much.  That means that in some cases they might well target high-ranking hubs/pages disproportionately, which is why high traffic hubs might go down, but lower scoring ones might go up?

         replyreport
         1. 97

            PaulGoodman67posted 12 years agoin reply to this

            By the way, I love all your analysis, Greekgeek - here and at Squidoo!

            replyreport
      2. 85

         Marisa Wrightposted 12 years agoin reply to this

         I do recall a discussion about this back when the switch happened, and there was a quote from Google explaining that sub-domains which were authored by the main site owner were treated as part of the main site, whereas sub-domains with different authors were regarded as separate sites.  I can't find it now, I'm afraid. 
         
         Sp if HubPages had switched completely over to sub-domains, I think there would be no question about it.  After all, look at Wordpress.com and Blogger - both are set up using sub-domains for each user, and it's clear each blog is separate and individual blogs have no effect on each other.
         
         What muddies the waters somewhat on HP is that we do have related links from other sub-domains on every page of our own sub-domain.  Some people have suggested this is a bad thing because it breaks the autonomy of our own sub-domain - but others think it's a good thing because of the cross-promotion.  Who knows?

         replyreport
         1. 91

            Will Apseposted 12 years agoin reply to this

            I was probably a bit simplistic in replying to Greekgeek before. The evidence is that subdomains are pretty well insulated but then there could be increasing drag as more and more sub domains are hit by Panda.
            
            We rely mainly on internal links to our pages for our link juice. The more accounts that get hit, the more that links to our pages are devalued. A big increase in Panda hits across the domain could be enough to reduce our rankings, especially if older articles with page rank are disappearing into the pit.
            
            I have also wondered if linking to Panda hit sub domains effects us. Linking to bad neighborhood sites (eg porn, malicious code sites)certainly hits a page. I have no idea if a Panda-hit sub domain is seen as a bad neighborhood. Probably not. It might be worth looking at though.

            replyreport
9. 91

   Will Apseposted 12 years ago

   From above-
   
   "Hubpages is gambling that dividing our content into subdomains invites Panda to treat each subdomain separately, so we won't be penalized for junk posted by other Hubbers.
   
   I'm still waiting for some kind of statistical study proving this. It makes sense, but I'd love to see the numbers."
   
   I would go and look at quantcast. After the subdomain shift, traffic took off like a rocket. Most of the better writers saw a 2,3 or even 4 fold increase in traffic. Despite the recent losses I am still getting twice the traffic I used to get.
   
   That is proof enough for me.

   replyreport
10. 97

    PaulGoodman67posted 12 years ago

    My traffic is even worse this weekend and I thought last weekend was bad.  I really do hope this is a case of "ebb and flow" and not just a "long, slow, ebb"! 

    replyreport
    1. 87

       Stacie Lposted 12 years agoin reply to this

       i can relate;my traffic is the worst its ever been...

       replyreport
11. 78

    shai77posted 12 years ago

    I had about 500-600 views each day, now less than 400 .
    Hope to see better days.

    replyreport
12. 97

    PaulGoodman67posted 12 years ago

    Traffic slowly creeping up.  Fingers crossed.

    replyreport
    1. 74

       Max Daltonposted 12 years agoin reply to this

       Yeah. The weekdays are usually the slower time for my hubs and the traffic has been trickling up since it went through its natural drop on Monday. Typically, my traffic bottoms on Thursday/Friday and then starts rising significantly through Sunday night. I've seen this happen over the last couple of weeks, but it's more pronounced today.

       replyreport
       1. 97

          PaulGoodman67posted 12 years agoin reply to this

          You seem to be talking about normal weekly cycles.  I'm experienced enough and have enough statistical info at hand to be fairly sure that my rise would appear  to not be related to the normal weekly cycle.  Early days, though.

          replyreport
    2. 53

       janderson99posted 12 years agoin reply to this

       Traffic this weekend at all time low - also applies to my other sites.
       What's going on?
       Qantcast data (Mar 24) shows traffic below 500K for first time since Xmas day. Average weekly traffic has been falling steadily since mid-Feb.
       Not a good sign!
       
       

       replyreport
       1. 89

          wildernessposted 12 years agoin reply to this

          Hmm.  Looks much like my own graph, except Sunday is generally quite good, while it begins to fall on Tuesday.
          
          Same general decrease since mid Feb, though.  Glad I could be of help in assisting the general traffic graph in its slide.

          replyreport
          1. 97

             PaulGoodman67posted 12 years agoin reply to this

             Greekgeek reckons Panda was run again on Friday - she says so on another thread. 
             
             My traffic was up at the middle and end of the week, then dived on Friday night/Saturday morning, more than normal, and hasn't recovered, yet.

             replyreport
             1. 62

                CASE1WORKERposted 12 years agoin reply to this

                Same here- has rallied a little tonigt but not as much as normal!

                replyreport
          2. 75

             Shadesbreathposted 12 years agoin reply to this

             Me too. That's exactly what I'm experiencing.

             replyreport
             1. 74

                Max Daltonposted 12 years agoin reply to this

                I've noticed that some of my most popular hubs have been losing traffic over the weekend, which is usually the time traffic to my hubs picks up. Not cool.

                replyreport
13. 60

    Randy Godwinposted 12 years ago

    "Pay no attention to the man behind the curtain"!
    
    
    
    
                                   

    replyreport
14. 83

    viryaboposted 12 years ago

    And the drop seems to be across all my sites.
    Strange.

    replyreport
15. 91

    Will Apseposted 12 years ago

    Relache mentioned elsewhere that lengthening days and improving weather in the US reduce internet traffic at this time of year.

    replyreport
    1. 74

       Max Daltonposted 12 years agoin reply to this

       Good point, Will. I was wondering that. I've been spending a lot of time outside this week, and we broke out the lawnmower this afternoon. I sure hope she's right.

       replyreport
16. 91

    Will Apseposted 12 years ago

    That might be the problem. Too much lawn mowing!

    replyreport
    1. 53

       janderson99posted 12 years agoin reply to this

       
       
       IF YOU CAN'T BEAT 'EM, JOIN 'EM

       replyreport
       1. 62

          CASE1WORKERposted 12 years agoin reply to this

          Mr panda, can you mow my lawn please?

          replyreport
17. 74

    Max Daltonposted 12 years ago

    It's definitely more than a summer slowdown. I'll concede that Hubpages overall traffic does appear to plateu during the summer months, but my traffic has been working its way down the ladder since Friday, and I'm scared to find out where the floor on this is. I'm curious to see what HP has in store to combat this. Are they out of ammunition after unleashing subdomains or is there anything else they can do that won't upset Google?

    replyreport
    1. 60

       Randy Godwinposted 12 years agoin reply to this

       Yo Max!  I noticed many of the hubs with unrelated articles listed in the "Related Hubs" section on the page had one or more of your auto repair related hubs listed.  I just wondered if you saw a rise in your traffic to these hubs because of the recent glitch?
       
       
                                

       replyreport
18. 74

    Max Daltonposted 12 years ago

    Hi Randy,
    
    I only have a couple of auto hubs: the cabin air filter articles for my two cars and the other one about the iPod adapter in the civic hybrid. The two cabin air filter ones did see a very small jump in traffic today and I thought that was odd. It wasn't a needle-moving jump by any means, but those hubs don't get a lot of traffic, and it was enough to make me raise my eyebrow.
    
    Thanks for pointing this out, though. I figured something abnormal was going on, but it wasn't like the 125 hits to a couple of articles over a two-day stretch from Facebook that threw my numbers off in the short term.
    
    Best,
    
    Max.

    replyreport
    1. 60

       Randy Godwinposted 12 years agoin reply to this

       Thanks for responding, Max!  Some of us are trying to figure out what's screwed up with the related hubs section of our niches.
       
                                    
                                                  

       replyreport
19. 81

    Rik Ravadoposted 12 years ago

    Panda 3.4 is reported to have been run around last Friday. This may explain further drops in traffic in the last few days ..
    
    http://www.seroundtable.com/google-panda-34-14926.html

    replyreport