1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Split testing of vendor page – a case study

Discussion in 'Clickbank' started by mancar487, Feb 28, 2017.

  1. mancar487

    mancar487 Jr. VIP Jr. VIP

    Joined:
    Apr 12, 2012
    Messages:
    1,918
    Likes Received:
    2,153
    Occupation:
    Clickbank Platinum Vendor
    There’s an inside joke among my team mates that I am addicted to split-testing, for the reason that I am always asking them to test and re-test variations of our clickbank pages for optimization of conversion rates and ultimately profits. We indeed carry LOADS of split-tests I was thinking of starting to share some of the data that we get from our tests (if these are useful to anyone). Some of the data may be useful to affiliates and vendors alike (such as color themes, font styles, etc..) while others may be more relevant to fellow vendors.

    To start with I want to share data from a fairly simple split-test between two price points - $37 and $39; however, as you’ll see, it is crucial to look carefully at the data obtained when there is a whole funnel in place (rather than a single product).
    Our funnel for this particular clickbank product is as follows:

    Main product (priced at $37) + 1 upsell ($17) and 1 order bump (a clickbank feature – think of it as an add-on – priced at $3).

    Now, the $37 vs $39 had already been tested pre-launch almost 2 years ago, and the 37 had out-converted the 39 significantly.

    This time, however, we got the following results:
    $37 conversion rate: 9.42%
    $39 conversion rate: 9.44%


    Initially, this looked like a double win – a better conversion rate and $2 more profit on the $39. However, we noticed that during the split-test period, our average customer value had dropped significantly, a fact that did not tie in with better conversions and a higher price point. We therefore looked at the conversion data for the upsell and order bump, and saw the following:

    At $37, the upsell converted at 36% and the order bump at 31%
    At $39, the upsell converted at 13% and the order bump at 14%


    As you can see, though conversion rate of the $39 was slightly higher than the $37, and though the former resulted in $2 more in initial profits, this price point killed the conversion rate of both upsell and order bump, ultimately leading to an overall drop in profits.
    Needless to say, we’re sticking with the original $37!

    Not sure if this data is useful to anyone – do leave your comments/feedback and I’ll be posting more split-test case studies if there is interest.

    Mancar
     
    • Thanks Thanks x 8
  2. Karim

    Karim Newbie

    Joined:
    Feb 5, 2012
    Messages:
    28
    Likes Received:
    5
    Location:
    Tunisia
    To me a $37 followed by a $17 upsell, looks more in harmony than a $39 followed by a $17. What if the $39 sale was followed by a $19 upsell wouldn't that keep the buyer in a buying mode.
     
  3. spectrejoe

    spectrejoe Jr. VIP Jr. VIP

    Joined:
    Sep 25, 2013
    Messages:
    2,105
    Likes Received:
    439
    Home Page:
    Damn that's awesome.

    Will bookmark this for the font, color & so on case studies :D
     
  4. Bigstar20

    Bigstar20 Jr. VIP Jr. VIP

    Joined:
    May 22, 2014
    Messages:
    263
    Likes Received:
    90
    Location:
    Australia
    I'm always fascinated by CRO/split testing. Something as simple button color change can literally add thousands of dollars to your profits.

    Definitely interested in seeing more of your data!
     
    • Thanks Thanks x 1
  5. mancar487

    mancar487 Jr. VIP Jr. VIP

    Joined:
    Apr 12, 2012
    Messages:
    1,918
    Likes Received:
    2,153
    Occupation:
    Clickbank Platinum Vendor
    That a very valid point @Karim! We had actually tested a $19 upsell price point on both a 37 and 39 initial item price in the prelaunch stage and the $19 upsell did poorly in both cases.

    Glad this type of data is of interest to a few of you :) Will post another case study soon.

    Mancar
     
  6. gman777

    gman777 Jr. VIP Jr. VIP

    Joined:
    Apr 7, 2016
    Messages:
    643
    Likes Received:
    490
    @mancar487 What do you think it could cause such a drop in earnings over time?

    Could it be that the people attracted by this $39 price point are different than the people that would normally get the $37?

    Or are they the same people and it has to do with some kind of psychological trigger?

    Cuz if they are different, then you may find a way to split them into 2 groups and sell them individually which would skyrocket your earnings.

    What do you think?