How Do You Run A/B Test?

12 replies
Hi Guys

I'm struggling with this whole A/B split testing thing..

I know I can do this in google analytics but for some reason it does not want to work...


So my question is what softwares do you guys use to A/B split test your pages?

Are there any out there that are cost effective and user friendly?

and just for fun

How do you guys make sure you get the most out of these kinds of stats?



Looking forward to your replies
#a or b #run
  • Profile picture of the author PopeMarketing
    I personally use Clickmagick for all my tracking/testing requirements.

    Generally speaking it's an awesome service, that does everything you'd need it to and more. Has the occasional hiccup, but nothing drastic.

    Hope this helps
    Signature
    For The Highest Quality Traffic, Free Sales Funnel And List-Building Consultation, As Well As Free Video Training (No Opt-in),

    Visit www.PopeSolos.com. Results Guaranteed.
    {{ DiscussionBoard.errors[10532673].message }}
  • i use cpv lab it does ab split testing as well
    Signature

    Join Next Live Mastermind Zoominar 100% Real World Secrets to Get Up And Running. Are you Stuck? Don’t miss it www.MonthlyMastermind.org
    {{ DiscussionBoard.errors[10532736].message }}
  • Profile picture of the author Marty Foley
    Hi Mike Vietri.

    If you have a WordPress site and want to start A/B testing just two pages against one another, I suggest you try Simple Page Tester: https://wordpress.org/plugins/simple-page-tester/

    A simple way to get started with it is to use another simple WP plugin called Duplicate Post to make a clone of your original page - available at: https://wordpress.org/plugins/duplicate-post/ , then make the change you want to test on the new cloned page, then set both up as test pages in Simple Page Tester, and let 'em run.

    Note that the free version of Simple Page Tester does not have conversion tracking built in, but that can often be done in other ways, depending on what you consider a conversion. I sometimes set up Google Analytics conversion goals in conjunction with tests using the tool, and assign arbitrary monetary values to goals, so I can measure by goal conversion rates and page value metrics in Google Analytics.

    If you want to go for more advanced A/B or multivariate testing at a higher price tag, try a tool like A/B Testing Software & A/B Testing Tools by Convert.com or https://vwo.com .

    Or if you're doing a simple test by rotating blocks of text / code on a single page, you can use something like: https://wordpress.org/plugins/ad-inserter/ . You can use it to test different ads, etc. I'm currently running such a test on one of my sites to get a feel for test the best title for an upcoming product.

    Marty Foley ~ PPC Traffic & Conversion Mad Scientist
    {{ DiscussionBoard.errors[10532755].message }}
    • Profile picture of the author MikeVietri
      Originally Posted by Marty Foley View Post

      Hi Mike Vietri.

      If you have a WordPress site and want to start A/B testing just two pages against one another, I suggest you try Simple Page Tester: https://wordpress.org/plugins/simple-page-tester/

      This is extractly what I needed,


      Thank you so much for your detailed reply!

      It saved me a lot of time... LOL
      {{ DiscussionBoard.errors[10533439].message }}
  • Profile picture of the author Mattdawg841
    That's great that you're ready to split test! The tracking program that I like the best and that has been the most user friendly for me is Clickmeter (about $20-30/month and totally worth it!).

    For split testing, they make it super easy as you can simply create a rotator link (all the pages you want to split test in 1 link and it rotates between them all) and drive traffic to that one rotator link.

    To get the most out of split testing, make 2 or 3 pages that are the same. Don't change 1 at all and that becomes the control. Tweek the second and/or third pages just a little (for example, add a sales video, change some colors, capitalize some letters, etc.)

    Drive about 100 clicks to each of the pages being split test and see which one converts better (all tracked by click meter). If the 'test' page converts better, that becomes the new control, make the same change on the other page and tweak something else on it. Over time you'll be able to maximize your conversions/sales or whatever it is the sites are trying to get!

    I hope that helps and good luck!
    {{ DiscussionBoard.errors[10532873].message }}
    • Profile picture of the author MikeVietri
      Originally Posted by Mattdawg841 View Post

      That's great that you're ready to split test! The tracking program that I like the best and that has been the most user friendly for me is Clickmeter (about $20-30/month and totally worth it!).
      Hi Mattdawg (Cool name btw..)

      Thanks for your awesome reply, and I know I'm super excited!!

      I've been working for quiet a while to get to this point.

      Currently, I only need to do very simple test, Just one page against the other.

      But I'm sure as my traffic and online activity grows,
      that my testing will become more complex and elaborate

      So thank you very much for your reply, I know that it will serve me well moving forward
      {{ DiscussionBoard.errors[10533442].message }}
      • Profile picture of the author JohnMcCabe
        Mike, Mattdawg gave you the basic idea, but I'd make a couple of changes.

        > Instead of making small changes, in the beginning you want to look for the home runs. Too many people start obsessing over things like button color or what shade of red to make a headline, or nonsense like that.

        You want to test variations of your offer (50% off purchase of 2 vs. BOGO [Buy One, Get One Free]). Test different headlines, and make them very different. Matt's idea of testing with and without video fits here. Bottom line, when you first start the process, you won't learn much from tiny tweaks.

        > You want to look at more than just 100 clicks. Unless, of course, that's the conversion you're looking for.

        Look for 100 results. For example, if you're building a list, run the test until you get 100 subscriptions. Getting 100 subscriptions on one of the offers is even better. Then use one of the online tools for measuring statistical significance. You want to make sure that the difference in the results means something. Did the winner beat the loser because of the difference in the pages, or was it random chance and you need to extend the test?

        > Just thought of one more thing. Be careful you don't get so deep into the testing that you forget WHY you're running the tests. At some point in your testing, you will reach a point of diminishing returns. After each test, ask yourself "will I get more bang for my buck by continuing to test this, or is there a better use for my efforts?"
        {{ DiscussionBoard.errors[10533601].message }}
  • Profile picture of the author Augustinus
    You can use different tracking softwares like voluum or clickmagic or qualityclickcontrol or simply create two different pages in wordpress or in HTML/CSS and send same amount traffic to them from same traffic source.

    This should help you determine a winner, you can split test totally different landing pages or just some elements on landing pages like color of subscription box.
    Signature
    {{ DiscussionBoard.errors[10532891].message }}
  • Profile picture of the author velvet
    Hi MikeVietri,
    This tool is very powerful, and will split test as well.

    You do want to know about the quality of your traffic too?

    ClickMagick - Track And Optimize All Your Marketing, All In One Place

    cheers, Mal.
    Signature

    Here is a Free Gift for you! :>) It's a great list builder too! https://velvetmice.com/vlander-wf

    {{ DiscussionBoard.errors[10537829].message }}
  • Profile picture of the author JC Web
    I use CPV Lab. You can split test multiple layers of your funnel simultaneously if you want. It does cost money though, but the usefulness is worth the price in my opinion. And if you are using paid traffic, it will pay for itself many times over very quickly. Any decent tracker should do split testing though. Since you should definitely be using a tracker anyway, it would be the easiest way to split test.
    {{ DiscussionBoard.errors[10537838].message }}
  • Profile picture of the author davidantonio
    I tried many platforms, I chose Dynamic Hits

    full with tracking and optimization features, very competitive price
    {{ DiscussionBoard.errors[10561738].message }}

Trending Topics