DISQUS

DISQUS Hello! 20bits is using DISQUS, a powerful comment system, to manage its comments. Learn more.

Community Page

20bits

Jump to original thread »
Author

An Introduction to A/B Testing | 20bits

Started by Jesse Farmer · 7 months ago

No excerpt available. Jump to website »

7 comments

  • Douglas Vos

    9 months ago

    I like the direction you are going in with this article... keep it going.
    Andrew King covers similar topic in his new book called Website Optimization.
    reply
  • Jesse

    9 months ago

    Douglas,

    Thanks for the kind words. Will do!
    reply
  • Prakash S

    8 months ago

    Does google analytics/ other analytics tool help you with A/B testing? Good intro to A/B testing.
    reply
  • Jesse

    8 months ago

    Prakash,

    Google has the Google Website Optimizer, which supports both A/B and multivariate testing, but it's fairly limited. It only measures "conversion" events.

    You can't, say, segment Google Analytics reports based on GWO experiments.
    reply
  • Frank Licht

    8 months ago

    Nice article :) Really helped me with the basics. You allready stated the Google Website Optizer is farly limited, but which tool would you recommend then? Thanks in advance!
    reply
  • Jesse

    8 months ago

    Frank,

    Give GWO a try, it might do what you need. I usually write my own.
    reply
  • Tim Cederman

    8 months ago

    Great article as always (and the preamble and follow up ones). As someone with a qualitative research background, I've really come around to the importance of data-driven design and taking the bias out of design decisions. However, while you qualified your statement "the landing page with the higher conversion rate is better", I think it's more than just unsustainable incentives (and the like) that need to be taken into account.

    Two pages may yield the exact same conversion rate - however they may offer a vastly different quality of the conversion (demographic of the user) and experience for the user. I'm not sure why there is often a rift between qualitative and quantitative research as they are really powerful when used in a complementary fashion. Without being able to actually talk to the user and getting a sense of who they are and what their experience was like it's difficult to infer this information otherwise.

    Looking forward to the next two articles!
    reply

Add New Comment

Returning? Login