System to 'exempt' people from 'pending' for 24 hours.

1. 82

   SimeyCposted 11 years ago

   There is a lot of anger and/or frustration relating to the 'pending' system and most of it is down to the fact that it's not clear why there are exceptions - this is justified and there should be open-ness. (Note this only is aimed at new hubs, not hubs that become un-featured later)
   
   One way to do this would be to produce a set of rules that guarantees someone becomes exempt. Such a scheme could be:
   
   -For all hubs created by the author in the last 90 days, calculate an average QAP score.
   -The pending time will depend on the average QAP score - so someone with a very high average would still have to go through the QAP process but wouldn't be pending.
   -This is a rolling score so as an author creates new hubs, if their quality dips then so does their average.
   
   This isn't a perfect scheme I understand - but I know that as a writer on here, knowing that if I continue to write decent hubs that I will be rewarded - then it will help me focus.
   
   Of course there have to be rules for new members - perhaps not getting an average until they've published ten hubs and automatically making them wait 24 hours?

   replyreport
   1. 96

      wildernessposted 11 years agoin reply to this

      Might work, and sounds fairly good to me.  Just a couple of things, though -
      
      Nubes should only have to wait 24 hours or until through the QAP, whichever comes first.  I'm sure that was your intent, but should be clear.
      
      You will still have the problem that if even one crappy hub gets featured there will still be a huge outcry that the entire system is faulty and needs scrapped.  Personally I can live with both but some will have a problem there.

      replyreport
   2. 60

      MelissaBarrettposted 11 years agoin reply to this

      I know it wasn't addressed to me but I'd like to add my two cents...
      
      I'd still make sure that a new hub went through QAP even if it was immediately featured.  If it fails then the hubber loses the right to immediate publication. 
      
      I've seen really bad hubs written by established hubbers with a decent tract record.

      replyreport
      1. 96

         wildernessposted 11 years agoin reply to this

         As I read it, that was a part of the proposal, and I would certainly agree with it.  Quickly featured or not, every new hub needs to go through QAP.
         
         I like Simey's idea of a rolling average better, though.  It's one thing to take away the right to bypass pending from a really bad hub, it's another to take it away from missing the QAP score by a point for something simple.  If the scores are bad enough somebody writing junk won't keep the right for very long anyway.

         replyreport
   3. 76

      AMFredenburgposted 11 years agoin reply to this

      The last Hub I published (I've only done 4 so far) got published in around 18 hours, which was a surprise to me. I had had the thing in progress for more than a month, and poked and prodded at it periodically, and was having views show up even though it was unpublished, so maybe someone at HubPages was peeking at it occasionally before it got published. I have no explanation for any of it; the only thing I can think is that the amount of time it takes to get published depends on when someone at HubPages can get to it; if the whole system is backed up, it probably takes the whole 24 hours.

      replyreport
      1. 92

         Judi Beeposted 11 years agoin reply to this

         The views before it is published are you going in to edit it.

         replyreport
2. 0

   Jane Holmesposted 11 years ago

   I'm fairly new on HubPages but have been satisfied with results to date. I agree that the hubs should go through the QAP. I, for one, don't want to publish a crappy hub. If its not up to standard, I need to pull it down and fix it. And, I think we can all have a bad writing day. So far the system seems to be working well for me.

   replyreport
3. 60

   MelissaBarrettposted 11 years ago

   Bumping.

   replyreport
4. 79

   derek gulbransonposted 11 years ago

   The pending time is not an arbitrary thing that we can just adjust. Most the the frustration that I've heard from people about it is due to the concern that if Google crawls it while it's has a NOINDEX tag on it, that this could adversely effect the speed or frequency at which Google might crawl it again. Is that the concern your speaking of?  You didn't really elaborate on the "anger and/or frustration" you were talking about.

   replyreport
   1. 60

      MelissaBarrettposted 11 years agoin reply to this

      I think the issue now is more about why some seem to get a fast-track through the QAP system.  Could you explain more about if that's true and how the process works?
      
      And if you could explain the timing of the QAP process maybe.  Like how long it takes an average hub to go through QAP and what factors influence that time?

      replyreport
   2. 78

      psycheskinnerposted 11 years agoin reply to this

      It relates to how some hubbers bypass QAP entirely.

      replyreport
   3. 82

      SimeyCposted 11 years agoin reply to this

      Derek - I have seen at least a dozen threads over the last few months about people getting fast-tracked, while other more senior members don't. Today is the first time that I've seen a conclusive post (from Paul) that goes a long way to explaining why this is happening - it may not answer every question everone wants answering - but it does give a reason....
      
      ...the anger has been building for over a year and if you hadn't noticed quite a few senior members have thrown up their arms and left.....

      replyreport
5. STAFF

   Paul Edmondsonposted 11 years ago

   There are a small number of Hubbers that go through.  It's a test.  We can collect data and see how they do compared to Hubs going through the QAP.  We have other tests running as well.  Testing is pretty fundamental for us.  Small groups of people usually make up a test group.  Sometimes we ask people.  Other times they get picked for various reasons.  It's something we will continue to do. 
   
   The process works at a high level like this.  Hub is published.  It gets rated by our internal algorithms.  It gets rated via the Hub Hopper.  At different time periods they are put through our dupe checker.
   
   We see that people feel it's important to be featured quickly, so we are working on it.  We've learned a lot since since we started the QAP process.  It will continue to evolve.  Ideally, I'd like a system that identifies content that people like, find useful or get enjoyment out of.
   
   To me, the best of HubPages is all the different voices.  Take Mark Ewbie with his unique style and compare him to Drax.  In my opinion, they are both great, but very unique.  The ideal system for us is one that features these voices, but separates it from stuff that isn't good.  My point is we want to be flexible enough to find the unique voices, but discerning enough to weed out the junk.  The problem is incredibly complex.  Some people have really great things to say, but have a hard time writing well.  Others write well, but aren't as interested in being helpful or useful. 
   
   Through different parts of the process there are trade-offs.  Where should the bar be drawn? How sure does the system need to be to make a decision?  How much to spend on each Hub?  What's the overall benefit?
   
   In the last few months, my feeling is Google has made more significant changes that impact sites like ours.  More weight on authorship.  I suspect they'll work on verifying credentials (algorithmically or through a service).

   replyreport
6. 82

   SimeyCposted 11 years ago

   Paul - can I make a suggestion - you, as co-founder and CEO should do a CEOs blog weekly. In it you could briefly cover what HP are doing (in broad terms like you just did) - you can let us know some of your thoughts about how it is helping ( or not) etc. Admitting that some things aren't working is just as important as letting us know things are working!
   
   Having some information from one 'face' that is consistent and tells us that you're not simply doing random things would at least remove some of the frustration and anger - you will still get people telling you that you are wrong - but no one can accuse HP of simply doing things for the sake of them - and you may find the open-ness may actually win over some of the harshest critics.

   replyreport
   1. 84

      rebekahELLEposted 11 years agoin reply to this

      While I think this suggestion merits some thought, I understand how he can't divulge too much information about the internal mechanics of how the site is run. ( I'm actually somewhat surprised with some of the current threads cropping up in the forums.)
      
      I think what would be great is something like SEOmoz does with their Whiteboard Friday, helpful, informative, short how-to videos about SEO related topics.  I've even taken the whiteboard Friday concept into my classroom.  The kids love it because it's visual.

      replyreport
      1. 82

         SimeyCposted 11 years agoin reply to this

         I agree there shouldn't be specifics - sometimes a statement like "We're encoraged by what we are seeing with our test groups" or even "Well that test group showed us xyz didn't work" - it simply conveying that some thought is going on even if we don't know what it is!
         
         I work for a large corporation and we get daily updates on what is happening in the business even when there are things to be concerned about - for instance - HP probably are more aware about Google changes than most hubbers are - so a weekly message could include - "look out for some ups and downs next week due to Google tinkering".
         
         It's nothing in detail, just helping us keep on the ball...

         replyreport
7. 61

   Randy Godwinposted 11 years ago

   I'd simply like to know if the idle/feature system is doing any good at all.  Has HP's traffic increased significantly since the program started or not?

   replyreport
   1. 85

      Rock_njposted 11 years agoin reply to this

      The stats regarding traffic have not been good since the idle program was put into place.  The idle program has caused many good writers to leave HubPages and move their work elsewhere.  It has also caused less in the way of news story hubs.  Any way I have measured it, the idle program has been bad.  It was sold as a way to improve hub quality, but now HubPages admits that quality concerns are not what idles hubs, it is actually a lack of traffic that idles hubs.  That much has been obvious to anyone who has seen their hubs idled (some very well written ones get idled, while the more popular ones get a pass regardless of quality).

      replyreport
      1. 82

         SimeyCposted 11 years agoin reply to this

         There really isn't enough data to support any correlation between stats and changes - this is from my knowledge of statistics and data analysis not from knowing insider information. I agree that parts of the program focus on different things now. I think what Randy would like is an official note from HP stating that "early stats indicate......"

         replyreport
         1. 85

            Rock_njposted 11 years agoin reply to this

            We can all examine the various traffic stats for HubPages at look at what has happened since the idle program was put into place.  Given the flat to decreasing long term traffic trend and the loss of many Hubbers since that controversial policy was put into place, I would say at this early juncture it has been a failure.  I have never seen anything that HubPages has done until this idle policy that has writers stating that they are moving hubs elsewhere and deleting their accounts.

            replyreport
            1. 60

               MelissaBarrettposted 11 years agoin reply to this

               I saw it with the subdomains.... I saw it after panda...  I saw it after the summary...Hell that one chick left because they changed the hubtool... Anything they do is going to make someone pack up their ball and go home.

               replyreport
            2. 82

               SimeyCposted 11 years agoin reply to this

               There are simply too many factors in play, and not enough time to validate any of them. If HP said that traffic improvement was down to their changes I'd also disagree. At this stage the traffic of HP varies based on at least three things - seasonal variations, hubpages changes, google changes.
               
               There are other intangibles that cannot be factored - Bing's increase in views, the loss of talented writers etc. 
               
               I've seen many writers leave, I've also seen many writers say they have had improved stats - again there still isn't enough data to give a true reading either way. I'm not saying this as a HP fan - in fact I'm being objective and using what I know from the types of analysis I've performed in my real work....

               replyreport
   2. 61

      Randy Godwinposted 11 years agoin reply to this

      I suppose my question wasn't clear enough?

      replyreport
      1. 55

         whoisitposted 11 years agoin reply to this

         No, your question was ignored.

         replyreport
         1. 61

            Randy Godwinposted 11 years agoin reply to this

            Most of them are! 

            replyreport
8. STAFF

   Paul Edmondsonposted 11 years ago

   I actually posted a very similar response in the past.  Maybe more than once (can't remember for sure).
   
   It's a good question on how to best respond to things that get asked more than once.    Largely, the community does a good job.  Although, sometimes forum threads take on a life of their own:)

   replyreport
9. STAFF

   Paul Edmondsonposted 11 years ago

   @Randy Godwin The QAP system is part of the quality efforts.  Starting with our move to subdomains and they will continue through the refinement of QAP and the moderation system. 
   
   There are a couple of fundamental questions.  First, do all authors experience similar traffic trends.  That's a resounding No.  Some people go up, others go down.  I think Google will continue to do things around authorship, but the first effort was to find a way to not have everyone treated the same way by search engines.
   
   Second, we've been told we have quality issues (here are the major moves).  We started with subdomains in 2011.  Then in August 2012 we started selecting Hubs to feature (some quality and some traffic/engagement).  That effectively reduced the number of pages in the site hierarchy.  Google's been digesting this change very slowly.  It's a bit curious to us why it's taken so long for them to update their index to reflect the change.  My estimate is this is about 50% reflected in Google's index (5 months later).  We see Google slowly dropping the pages. 
   
   Now to the rating part.  We are focusing on new Hubs first.  The goal is to do a better at keeping low quality Hubs from ever going into the index.  I still think we have a ways to go on the front door.  We don't have enough ratings to make the decisions we need to on all new Hubs, so we are working on getting more raters.  Also we are looking at ways of doing it more efficiently.  Then we will move to the backlog. 
   
   I think it will take several more months for these changes to work through.  The only metric we can be sure of is that with the QAP system, the average quality of a featured Hub has increased. 
   
   We will also shortly have the ability to benchmark ourselves against other sites.  For example we can look at the corpus by author, category etc and see what the quality distribution is compared to other sites.  That should help us in tuning our system.

   replyreport
   1. 82

      Marcy Goodfleischposted 11 years agoin reply to this

      Thank you, so much, for explaining the approach you are taking at the Admin level.  I'm very much in favor of riffing the site of the numerous duds we know are here. It's interesting to see that Google has actually dropped some of these, but that it took several months.
      
      I'm not sure if you read my post (somewhere - it was a long night last night - not sure where it was) that my traffic has grown considerably since the idling process started. I think a piece of it is my new awareness of the need to watch monthly views for each hub, look at the age of the hub, check when it was last updated and try to preempt any idling that might be pending.
      
      In many instances, I have seen really weak subheads - so I tweak them with more search-friendly terms. And I am sure I could go in several months from now and see yet more ways to improve things.
      
      At any rate - aside from the tanking many of us saw in late September, my traffic is 50-60% or more higher than it was before the late September attack by Google. Or whatever that blip was.

      replyreport
   2. 61

      Randy Godwinposted 11 years agoin reply to this

      So the answer to my question is, you can't tell yet if the program is working or not?

      replyreport
10. STAFF

    Paul Edmondsonposted 11 years ago

    We can tell quality is improving on the site.

    replyreport
    1. 85

       Rock_njposted 11 years agoin reply to this

       That kind of depends on how you define quality.  If having more hubs about popular topics that have keywords used in a way that causes them to rank higher in Google is your measure of quality, then I guess it's working.  But I have seen a decrease in creative writing and off-beat topic hubs on HubPages since the idle policy was put into place.  I considered many of these pieces to be quality pieces of writing.  I have also seen the number of Hubbers decrease dramatically over the past six months and many Hubbers cite the idle policy as their reason for leaving.  Also, HubPages has stated that it is traffic that actually causes a hub to go idle.  A poorly written hub can get a lot of traffic and not get idled, so I'm not sure about the improvement in quality due to this policy.
       
       Needless to say, I am not a fan of the idle policy.  I think a warning system would be a fair way to implement it.

       replyreport
       1. 61

          Randy Godwinposted 11 years agoin reply to this

          Here you go! 
          
          
          https://www.quantcast.com/hubpages.com

          replyreport
11. 61

    Randy Godwinposted 11 years ago

    The voice of experience!

    replyreport
12. 54

    janderson99posted 11 years ago

    Traffic is King! Quality can never reign supreme, while the old crap remains! Long Live Traffic!
    
    

    replyreport
13. 96

    agilitymachposted 11 years ago

    I don't get why people are ticked that their hubs are being featured in 24 hours while other are being featured immediately.  Is it ego?  Are people that sensitive?
    
    I don't care if it takes 24 hours for me to get a hub featured as long as it gets featured (and all of my hubs have passed QAP).  I don't mind that HubPages is pushing for quality, information-filled articles for SEO.  If you don't want to write like that, go elsewhere.
    
    When I use Google, I'm looking for fast information.  Not creative writing.  Not poetry.  I'm looking for how to fix my plumbing or how to fix my car.  I'm not looking for the latest chapter of your book.  I have several creative writing hubs that became idle, and I've let them remain so.  They won't get Google traffic, and they are really not what this site is about.  I keep them for me and those that wish to search for more of my writing.  That is all.  I don't gripe about it though.
    
    When one of my Hubs is featured, I go to my webtools and "fetch" a Google bot to search it.  Then, poof, it's indexed and ready to go.  My ego isn't so sensitive that I expect to be published immediately because I'm all that and two bags of chips. 
    
    So why are people so up in arms?

    replyreport
    1. 96

       wildernessposted 11 years agoin reply to this

       There have been some reports that requesting a hub be indexed doesn't always work.  The thinking seems to be that google "sees" the noindex tag and goes away.  For weeks sometimes, and the fetch bot won't come even if you request it.
       
       That means that any google traffic could potentially be set back several weeks; a tremendous blow to current event type hubs and a lesser one to more evergreen topics.

       replyreport
       1. 96

          agilitymachposted 11 years agoin reply to this

          The fetch has worked for me, but I can see the issue for current event hubs.

          replyreport
       2. 82

          Marcy Goodfleischposted 11 years agoin reply to this

          I'm okay, generally, with a waiting period before something goes live.  But if the mechanism sets up a counter-productive message to Google, that's not so great.
          
          SUGGESTION TO ADMIN:  For those Hubbers not in the 'automatic pipeline,' could the process be one where you submit it 'pre-publication,' and Google doesn't even know it exists (so they can't see a No Index thing).  Then, after it is vetted, the system automatically publishes it?  There would still be a waiting period, and a vetting process, but no negative effect from a No Index stage.  At least I think that's how it could work?

          replyreport
          1. 85

             Rock_njposted 11 years agoin reply to this

             That's a good suggestion Marcy.  Since we don't actually know how Google works internally, but we know that the NOINDEX tag is toxic in Google for any page that has it and that toxicity may live on after the NOINDEX tag has been removed, it would be best to avoid slapping the NOINDEX tag on new hubs under internal review at HubPages.  Perhaps they could be placed in a hidden directory or something while under review, so Google isn't aware of them.

             replyreport
             1. 82

                Marcy Goodfleischposted 11 years agoin reply to this

                For that matter - if we have hubs that 'go idle' for whatever reason, why couldn't they be administratively unpublished, with a flag on our accounts page to check them out and address whatever issues are there.
                
                It seems the No Index (or however it's styled - all caps?) is not only demoralizing to writers, but counter-productive to the site, as well.  A staging area would get rid of most of those problems (a pre-publication, holding zone).

                replyreport
                1. 96

                   wildernessposted 11 years agoin reply to this

                   If a hub becomes unpublished all links pointing to them are suddenly invalid and readers cannot see the hub at all.
                   
                   Backlinks you (or others) have put in other hubs, on FB or in other forums must now all be removed.  Particularly those here on HP - when I unpublish a hub I'll generally have a half dozen or so backlinks to track down and remove.
                   
                   A pre-publication, holding zone, is probably unworkable for a similar reason.  I'm no programmer, but have a hub with a URL that cannot be seen means that HP must be able to recognize incoming requests for that hub and provide the proper URL without letting google see what they're doing.  Sounds difficult.
                   
                   Probably better and far easier to simply not publish a hub until the vetting process is done.  That will still negatively effect current event hubs but seems a better solution than letting a google bot see a noindex tag.

                   replyreport
                2. 85

                   Rock_njposted 11 years agoin reply to this

                   I heard that HubPages is working on some kind of warning system for Hubs that are nearing the NOINDEX criteria, so Hubbers have time to make changes before the Hub is officially unFeatured and has a NOINDEX tag slapped on it.  That seems like a fair solution to this issue that seems to be the subject of so much angst and debate here at HubPages.

                   replyreport
                   1. 82

                      Marcy Goodfleischposted 11 years agoin reply to this

                      I know that's been brought up in the forums.  I hope it's on the to-do list.  It appears the filtering system for No Index covers every hub on the site, whether it's one that's bad and has languished for three years, or a new one, of good quality, thar just hasn't gained traction yet.  I'm sure all of that is difficult to program, so I hope the ideas here help the Admins come up with ways to adjust things.

                      replyreport
                      1. 96

                         wildernessposted 11 years agoin reply to this

                         I don't know that it's on a to-do, but it is on the to-look-at list.  As several have commented, the programming is very difficult and not an easy task at all - the return must always be balanced against the cost.
                         
                         Apparently the primary thrust right now is new hubs - keep the junk off of HP.  Finding a way to do that that can't be gamed or bypassed isn't easy (we've all seen obvious bugs) but it can hopefully be done.  When it is, then the primary thrust will be to attack the older trash and work on that but it's going to take time, and a lot of it.

                         replyreport
                         1. 85

                            Rock_njposted 11 years agoin reply to this

                            In a recent forum post, I asked Simone about about a modification to the current unFeature / NOINDEX, so that a warning is given rather than a NOINDEX tag out of the blue.  She said it is being discussed at HubPages and is a top priority, and also noted that it would be difficult to implement.  I think this issue is a top priorty because they are literally losing Hubbers over it, and not all of them are the low quality writers and hubs that they were hoping to wipe from their system with this idle / unFeature / NOINDEX policy.  I'm not sure if their goals are being met due to the way this policy was implemented.

                            replyreport
    2. 85

       Rock_njposted 11 years agoin reply to this

       Someone was recently ticked off about the 24 pending policy because a brand new Hubber that copied original content had a hub featured quickly, in well under 24 hours, while he and other established hubbers have to wait and have the NOINDEX tag on their hub.  I think everyone can understand why this would upset people who have worked hard for years at HubPages to see such a breakdown in the 24 hour waiting period process.  HubPages says their review software wasn't working correctly over the weekend and the offending hub has been removed.

       replyreport