SUGGESTION: Features of High Traffic Hubs

1. 55

   janderson99posted 12 years ago

   In a previous post, Paul E asked for statistical summaries that people may be interested in. It would be very helpful to have an idea of the features that correlate with high traffic. Here is my suggestion - perhaps the results could be reported in a blog.
   Features of High Traffic Hubs
   Filter the hubs database for hubs with the highest traffic over the last 30 days, and rank by traffic. Present data for the following 3 groups:
   # 80-95% best performers (eliminate the top 5% - probably "lottery winners", viral etc.
   # 65-80% best performers
   # 50-65% pest performers
   
   Present statistical summaries for:
   # age of hub
   # author's account age
   # number of followers
   # total number of words
   # video present
   # poll present
   # amazon capsule present
   # number of images
   # number of hubs in sub
   # category breakdown
   # number of links on the page
   # number of incoming links
   # average hub score

   replyreport
2. 0

   Arlene V. Pomaposted 12 years ago

   This is cool.  Thank you for your efforts.

   replyreport
3. 85

   K9keystrokesposted 12 years ago

   I think that when HP gave us the info on "Stellar Hubs" this info was provided, for the most part. However your suggestion seems to be a pretty good idea above and beyond this.
   Here is the link to Stellar Hubs hub found in the learning center: http://hubpages.com/learningcenter/Elem … tellar-Hub

   replyreport
   1. 55

      janderson99posted 12 years agoin reply to this

      I wonder whether this was based on statistics related to traffic?

      replyreport
4. STAFF

   Paul Edmondsonposted 12 years ago

   We need a bit of engineering help to get the data into a form we can use to do the full analysis you suggested, but I'm not sure that looking at hubs individually and correlating that data with other account aspects would be useful if we just look at the top trafficked hubs. 
   
   So, we could do it at the account level and take a random sample and use average views per hub as the metric, but it still won't account for all the reasons people do well or poorly.  Each time I look at this type of data, I need to actually read the hubs to get a sense of why they are doing better.  The reasons vary from they picked the perfect niche, the Hub has good links, the hub is excellent, to some factor we can't discern. 
   
   Keep the data ideas coming.  I'm hoping to have a resource to work on these shortly.

   replyreport
   1. 55

      janderson99posted 12 years agoin reply to this

      Thanks Paul, I know the limitations, but some bland stats will yield valuable insights and may perhaps dispell some of the myths. Its a changing feast, with the 2 G P's Having a stats system in place may help to 'roll with the puches' and 'go with the flow' as things change.
      For example the evergreen vs freshness debate.
      "article quality" vs seo for title, backlinks, interlinking

      replyreport