An experiment - The impact of making my hubs 'Stellar' on SEO

1. 76

   Paul Maplesdenposted 11 years ago

   I've decided to run an experiment to see how creating 'quality content' will impact on Search Engine Optimisation, performance in Google, traffic and search results.
   
   I will be refining my existing hubs using the quality guidelines as recommended by HP through the guidelines for stellar hubs: http://hubpages.com/learningcenter/Elem … tellar-Hub
   
   I will go about this as follows:
   
   1. I will track all of my hubs statistics 'as is' (prior to updating them to be 'stellar) for the past 30 days
   2. I will then go through each hub and ensure it has all of the 'stellar' qualities; this won't be a 'tick box' exercise, I genuinely want to make each hub as good as it can possibly be. However, as a minimum, each hub will have all the elements of a stellar hub as defined in the Learning Center article
   3. I will then track the results to see what the impact is of updating content over 30 more days
   4. I will analyse all of these results and write a hub (or more) on what I have discovered
   
   My plan is to see if there is a strong correlation between 'stellar hub' content and whether the quality of that content impacts on search positioning, click through, traffic and other metrics.
   
   I currently plan to measure the following areas (for 30 days before and after making the changes). I will measure this for each hub and as a total. (GA is Google Analytics measures, WMT is webmaster tools measures)
   
   - Amount of traffic - Overall (GA)
   - Amount of traffic - Search (GA)
   - Time on page (GA)
   - Bounce Rate (GA)
   - Search impressions (WMT)
   - Search clicks (WMT)
   - Click through rate from organic search (WMT)
   - Average hub position in search engine results pages (SERPs) (WMT)
   
   I know that there will be some factors beyond my control that may impact my results (Google Algo updates, competitor activity, natural improvement, authority etc.) but I am still hoping that the experiment will throw up some interesting results.
   
   I'll write some comprehensive hubs sharing what I have found as well.
   
   If anyone has any suggestions for other measures / metrics or other things I might want to incorporate, please let me know.
   
   I'll also share all of the raw data behind the experiment once it has concluded.

   replyreport
2. 84

   psycheskinnerposted 11 years ago

   I would suggest a control group of hubs you edit and update, but not using these recommendations. Just updating with cause a short term boost.

   replyreport
   1. 76

      Paul Maplesdenposted 11 years agoin reply to this

      Good point, I did think about a control group, but not one I had edited, that's a really good idea.

      replyreport
      1. 94

         sallybeaposted 11 years agoin reply to this

         I completely agree, just a little editing, changing of ads, tweeting, pinning etc., causes an immediate reaction!  I just love those little red arrows.

         replyreport
3. 94

   vocalcoachposted 11 years ago

   I'll be looking for those hubs revealing the statistics of your experiment.  If I find that you are indeed rewarded by your hard work, I'll follow suit.  Thanks Paul!

   replyreport
   1. 53

      janderson99posted 11 years agoin reply to this

      The Proof is in the Pudding
      If Google ranks Stellar hubs higher then you should see the Stellar ones at the top. I ran a simple Google search for a topic such as
      "potato recipe hubpages"  etc. [the "hubpages" simply gets pages from HP.]
      I counted the number of words, images, videos and tables.
      The results are shown below:
      Sure its a straw poll but it clearly shows that articles with small amounts of text, and few images and videos often appear higher in the SERPS
      ENJOY!
      
      

      replyreport
      1. 76

         Paul Maplesdenposted 11 years agoin reply to this

         That's a good approach too - I can see how that's a useful comparison and would be an interesting side piece to the experiement. Thanks for sharing the data.

         replyreport