Split Testing - Results?

9 replies
Hello fellow Warriors!

I know that some of you do it religiously, and others forgo the opportunity all together.

I have been split testing for some time, and I love it when I find a new element that enhances my conversion rates - but I also dislike it when I try something new that tanks.. haha. (I kind of wish I'd just gone with the good one I already had, y'know? )

With that said, I'll be split testing until the day I die - both on the internet and off - because it's just a good idea. However... Do you have any particular methods that you use to avoid testing the same thing multiple times?

For instance, I'm feeling like I have to test things for each individual niche even if they didn't work in another - and sometimes they don't work in either.. haha

But, I don't know that unless I try it out and see for myself.

If you're having the same issue and don't have a solution, just chime in so I know I'm not alone out there.. haha :p

Personally, I feel like the results of my split testing are a pretty personal topic ... however, that being said, I'd still love to get my hands on the results of OTHER people's split testing (who wouldn't?).

Is there a place where information like this is shared so I can see? If revealing my own data is part of the cost I'd probably be willing to do it.


Thanks in advance,
-Adrian
#results #split
  • Profile picture of the author Shaun OReilly
    I'm a testing and tracking nutjob too!

    Here are some testing-related resources that I really like:

    The Conversion Rate Experts site is awesome:

    Conversion Rate Optimization | Conversion Rate Experts (no affil.)

    The guys at Marketing Experiments share test results
    too:

    Discover Which Marketing Programs Really Work (no affil.)

    Visual Website Optimizer
    share some testing case studies
    from heavy hitters:

    A/B Split Testing | Multivariate Testing | Case Studies - Visual Website Optimizer (no affil.)

    Many people TALK about testing but very few Internet
    Marketers actually DO it.

    Surprising really considering that TRAFFIC and CONVERSION
    are such big elements in successful Internet Marketing.

    Dedicated to mutual success,

    Shaun
    Signature

    .

    {{ DiscussionBoard.errors[5986252].message }}
  • Profile picture of the author BIG Mike
    Banned
    [DELETED]
    {{ DiscussionBoard.errors[5986289].message }}
    • Profile picture of the author Shaun OReilly
      Originally Posted by BIG Mike View Post

      Whenever I read about something that someone has said works really, really good or bad, if I think it's applicable to what I'm doing, I'll run a series of tests to obtain data unique to what I'm doing.
      Absolutely agree.

      I observe other people's split-tests mainly to get ideas
      on what I could test myself.

      It's a great way to get left-of-field testing ideas and
      also get out of the IM dogma box on what supposedly
      'works' and what doesn't.

      For example, here's an interesting test result from
      GetResponse where they found that having TWO
      call-to-action buttons increased overall converisons:

      Whoa! Free trial button did not decrease paid signups, but increased trial signups by 158% (no affil.)

      IM dogma would say "just have one call-to-action"
      but that's not always true.

      That said, I still test it myself.

      I come up with a test and then ask my audience to
      vote on it with THEIR clicks and wallets.

      Dedicated to mutual success,

      Shaun
      Signature

      .

      {{ DiscussionBoard.errors[5986315].message }}
      • Profile picture of the author Adrian Int
        Originally Posted by Shaun OReilly View Post

        Many people TALK about testing but very few Internet
        Marketers actually DO it.

        Surprising really considering that TRAFFIC and CONVERSION
        are such big elements in successful Internet Marketing.
        Originally Posted by BIG Mike View Post

        Over the years I've seen a lot of marketers get sucked into believing that so and so said he tested that and the results were this...so they end up not doing something important or doing something unproductive.
        Sadly, I think you're right. I love it when I click on link, then suddenly I'm confronted with an offer like "Get my 99.9% converting squeeze page when you share your e-mail below!".

        I hope it goes without saying.. that number is ridiculous. I'd believe it - MAYBE - but only if I knew that it was several steps into the funnel before the audience was exposed to it. The only other scenario where numbers like that might pop up (at least, that I can think of) is if only one person landed on the page and happened to sign up.. haha

        Regardless, I think a lot of people get sucked into the idea that the squeeze page will deliver because the numbers on the page said so.. even though it might've been used for a completely different market niche.

        Originally Posted by BIG Mike View Post

        It's encouraging to see others going to the time and effort to test everything
        I agree. I knew there'd be at least a couple other out there with the sense to want some "hard facts" to help improve their bottom line.. haha

        Originally Posted by Shaun OReilly View Post

        Absolutely agree.

        I observe other people's split-tests mainly to get ideas
        on what I could test myself.

        It's a great way to get left-of-field testing ideas and
        also get out of the IM dogma box on what supposedly
        'works' and what doesn't.

        I come up with a test and then ask my audience to
        vote on it with their clicks and wallets.

        Dedicated to mutual success,

        Shaun
        I like it Shaun. I do this often, too.

        Sometimes the best place to learn isn't from some master in the art (topic), but instead from your competitors. They're the ones in the trenches with you, after all. I'll be honest.. I've snagged ideas from others in my niches before.. and I'm better off for it.


        Oh, and thanks for the resources, Shaun. I'll be checking them out after the kettle boils and I make some more tea.. haha. In my experience you always offer great value in your posts, and it's much appreciated.

        And something tells me that Mike has added some value once or twice before, too... Congrats on 10k posts, soon! :p


        Cheers,
        -Adrian.
        Signature
        ATTENTION: Improve your list building. Free report and mp3
        {{ DiscussionBoard.errors[5986421].message }}
  • Profile picture of the author JimDucharme
    There are two things I've learned about email marketing...

    It depends.
    &
    Test everything.

    I think Which Test Won is the site you are looking for sir. Split testing nirvana.

    Regards,
    jim
    {{ DiscussionBoard.errors[5987652].message }}
    • Profile picture of the author s1ngular1ty
      You guys done much dayparting?

      Split testing the best ad delivery times is the only way I can get Facebook ads really profitable these days.
      Signature
      Viral Marketing Machine ONLY $9! How to create your own viral, monetised, niche websites that AUTO PROMOTE themselves on Youtube, Facebook, Twitter and Google. NO SITE CONTENT NEEDED![B]
      {{ DiscussionBoard.errors[5987950].message }}
      • Profile picture of the author Adrian Int
        Originally Posted by s1ngular1ty View Post

        You guys done much dayparting?

        Split testing the best ad delivery times is the only way I can get Facebook ads really profitable these days.
        When I run FB ads, I do this as well. Your prospects are only online at particular times (and so too, should your ads be). Personally, I'll usually run at least 10 different yet similar ads to begin with, then whittle it down to a couple "winners" when I need to run them.

        I'm not sure how you're setting them up, but "landing" your clicks on a facebook page will help you keep the cost down, as will promoting it in a way that someone might actually hit "like" under the ad.

        Back to testing the timing of ad delivery...

        I do this with my lists, too. I get mixed results. I'll be better off for it long-term than if I never gave it a shot, though.. haha.

        Cheers,
        -Adrian
        Signature
        ATTENTION: Improve your list building. Free report and mp3
        {{ DiscussionBoard.errors[5988032].message }}
  • Profile picture of the author thetrafficguy
    Yes you can study others, and I suggest doing so. But you should only look at other's split tests results as a possibility and something to test.

    You NEED to test your stuff out. Split test religiously!!!
    Signature
    "SUPER AFFILIATE EXPOSES Highly Profitable Traffic Source!"

    "Use This Trick To Make An Extra $50 - $500 Per DAY!"
    Click Here To Check It Out Before The Price Jumps Up AGAIN!
    {{ DiscussionBoard.errors[5987991].message }}
  • Profile picture of the author JKflipflop
    A much needed thread here. Its always good to look at other people's results but despite everything that may be provided to you, it is still better to get into the habit of factual split testing yourself.

    Conversion rates can change even with the least sensitive changes to web pages at times. Just my 2 cents.
    {{ DiscussionBoard.errors[5988036].message }}
  • Profile picture of the author philipdwyer
    I am in real estate and the best way I learned to split test was to find out what the most successful real estate investor is doing and copy.
    Signature
    --
    Philip Dwyer
    {{ DiscussionBoard.errors[5988360].message }}

Trending Topics