5 {{ upvoteCount | shortNum }}
5 {{ upvoteCount | shortNum }}

I hate split testing

by smith33122
4 replies
  • SEO
    • |
So I have my site setup approx 1200 new visitors in the first month - not bad for b2b..

My video on the front page and a ctr of about 1.4% on the Adsense - I have now taken the video off and not a single click today, but I have to leave it for a week just to be sure.

I hate split testing
#hate #split
  • Profile picture of the author KyleGolemMedia
    While you may hate split testing, it says so much about our web properties. Hell, SEO, SEM, and Social Media can all be effected positively or negatively by moving a single thing around on your home page.

    Understandable that you aren't interested in the time it takes. Sounds like you need somebody like me doing the grunt of it .

    Good luck and hopefully the time you spend testing is worth it! In my experience, I always learn something new.

    Kyle
    Signature
    Need Help Growing Your Audience?
    Let an experienced Marketer help! PM me with project details.
    Google AdWords, Bing Ads, Facebook Ads, Twitter Ads, etc.
    {{ DiscussionBoard.errors[6459132].message }}
  • Profile picture of the author dburk
    Hi smith33122,

    What you have described is not truly split testing. A true A/B split test will split the traffic, typically 50/50 with half randomly going to your current page (control), and the other half going to the experimental page.

    The key concept being the control and experiment are tested concurrently, and not consecutively. You probably need to repeat the test a couple of times, if you are running consecutive tests, to confirm the accuracy of your consecutive tests.
    Signature

    * Get Results - Outsource Your PPC Management |
    *     Don Burk Advertising & Marketing - www.donburk.com

    {{ DiscussionBoard.errors[6459187].message }}
    • Profile picture of the author smith33122
      Originally Posted by dburk View Post

      Hi smith33122,

      What you have described is not truly split testing. A true A/B split test will split the traffic, typically 50/50 with half randomly going to your current page (control), and the other half going to the experimental page.

      The key concept being the control and experiment are tested concurrently, and not consecutively. You probably need to repeat the test a couple of times, if you are running consecutive tests, to confirm the accuracy of your consecutive tests.
      I agree and am conducting the test against several similar sites, but just hate to lose any revenue just to prove something I should know.

      What is interesting is just how much effect video has, basically i am telling the viewers what i want them to do and a percentage do I am now improving the quality of my videos to improve further still and am testing with pretty dumb girls, puppets and a zombie
      {{ DiscussionBoard.errors[6459955].message }}
  • Profile picture of the author smith33122
    What may interest people is that I used a script to create the site with 3,500 pages and no SEO and yet Google has not objected and whilst the content is very basic scraped stuff I put up downloadble pdf's which should prove of value to anybody interested in marketing
    {{ DiscussionBoard.errors[6459985].message }}

Trending Topics