If you want to know the “secret” to getting rich online, split testing is it.

However, it also might be the most “unsexy” part of the business.  It’s not fun to talk about like traffic or launch videos.  And it’s certainly not something you see any of the gurus teaching.

But if you want to make a bunch of money online, you MUST split test. It’s the biggest difference between the guys who are failing and the guys who are making a boatload of money online.

Split testing is really a “discipline” type thing.  And being a good Ohio boy with good work ethic, I’m pretty committed to testing everything.

So today I wanna open up my playbook and share with you the results of 2 BIG impact split tests I ran for The 31 Day Fat Loss Cure

Add To Cart Button Testing

So the first test I did was with different “Add To Cart” buttons on our site.  Buttons on your site are one of the MOST important things you can test for sales.

My friend, Jon who does all kinds of landing page optimization and conversion tracking for big companies kind of sparked this idea at the last Elite Retreat conference…

He got me thinking about our button and how to make it jump out more, so I got right to it and tested a new brighter button for the site… I also added some text above that says “Claim Your Copy Now”

These 2 little tweaks gave me a 78% increase in sales! Overall the return on this little change has been over $70,000!

Now the point of this is not to use the same button as me – the point is you MUST test all kinds of button colors and shapes if you want to get maximum conversions for your product or offer.

For starters I always recommend testing…

  • big buttons
  • orange buttons
  • eye catching buttons

In addition to the button testing, the other significant test I ran was with the header on our site…

Header Testing

The header on your website is a pretty big deal when it comes to conversions.  Most people royally mess this up and use a BIG header that takes up a lot of space.  From my experience, small headers usually do MUCH better conversion wise.

In this conversion test, I made our header a little more professional looking – and also added in the stock photo cartoons.  The use of cartoons is little trick that works really well for conversions.  Everyone loves cartoons, so putting cartoons on your site can be an easy way to instantly create trust with your visitors…

Here’s the split test…

As you can see, I got a 27% increase in conversions with this header change.

If you’re looking for something to test with your header, I would try the following…

  • smaller header
  • cartoons
  • good, clean logo
  • light in color like grey or light blue…

Overall these little conversion increases helped add about $93,000 to my pocket that I wouldn’t have had if we didn’t split test.

– Justin

By Justin Goff

Justin sells the highest paying fat loss product on Clickbank called the 31 Day Fat Loss Cure. Get more info on being an affiliate here or check out the best damn podcast about the crazy world of Internet Marketing

145 thoughts on “How 2 Split Tests Made Me Over $93,000”
  1. I love this stuff. I find testing so fascinating how seemingly small changes influence human behavior, particularly with that buy button. You would think it would not matter as much as it does, but it really does matter, even from my own experience with a few other sites of mine. Best though, to change a few small things at a time so you can track results and correlation.

    1. Amazing results for such small changes. Goes to show that every little thing matters and that you can’t just trust your instincts when it comes to what customers want.

    2. I usually move UI elements around and do split testing to see what’s the most effective. A simple step but often gets me good results.

  2. Thanks for the tips, Justin! I haven’t done split testing yet so I’ll make use of the strategies you use and see how it goes.

    1. Great post! I use the Google Website Optimizer to get really deep down and dirty with split testing whether its A/B testing or MV testing.

  3. Very interesting Justin.
    Split testing is crucial in affiliate marketing in media buys, landing pages, email marketing and more. It’s the difference between an regular affiliate and a super affiliate.

    1. I’d like to learn more about split testing. Can you recommend the best reads on the topic?

    2. Interesting read. Split testing seems to be a big task involving lots of time. But from the results you’ve had, it sure looks like something that’s going to be worth the time and money invested on it.

      1. It’s really not involved at all. Google’s web site optimizer is free and takes a few minutes to install for any one with basic html knowledge.

        Find an element you wish to test, like buttons. You can start simple and test different color options of your original button. You place those in the testing software and click go. Couple more steps but they are that easy.

    1. Yeah. Layout, design, and the overall look and feel I guess. That’s a lot of work especially for those who are just starting out like me. But I don’t mind working on those as long as they’re effective in increasing revenues

  4. I’ll be launching an online bookshop soon. When do you think is the best time to do split testing?

    1. Why not do it as soon as you’ve launched? That way you’ll have a fairly good idea as to what drives more traffic and go with it early on.

  5. We’re all after maximum effectiveness of our efforts. The only way eliminate to find out what landing page is generating the most conversions is to test it. That way you’ll get rid of those that underperform and focus on pages that are getting the most results.

    1. It’s a lot more profitable to put your time and resources on efforts that produce the best results. Why waste time on something that’s not working, right? This split testing idea is definitely worth a shot.

  6. I am more of a design and SEO guy but I have had clients who are into affiliate marketing. The tips you shared here would come handy next time I work on an affiliate’s website design. Thanks, man!

  7. I don’t know much about split testing but I’d like to give it a try. This just might be what I need to do now to get more conversions.

  8. I’ve heard of this technique but have never used it. Great post. You’ve outlined its relevance.

    1. Sometimes changing something that may look trivial can significantly improve conversions. Split testing is the best way to find out which changes yield the most impact.

      1. I agree. It’s only with split testing that you’d be able to assess the effectiveness of a page’s design or content.

    1. Try Google Website Optimizer. It’s the only free tool I know but there are probably more out there.

  9. I agree with what you said about big, eye-catching buttons. I always find myself checking those out when I see them on some of the sites I visit.

  10. This could be useful in checking if the layout or ad placement is effective or not. This would be the first time that I’d try split testing. I hope I’d be able to get some really good results from it.

  11. Nice post! It seems like a great strategy to unlock a website’s potential. Anyone who’s serious about improving conversions should give this a try.

  12. Split testing can help ensure that you are maximizing the highest converting version of your site possible. It’s not a simple task and almost always entail a long process but the potential rewards are massive.

  13. What you like on your site may be very different from what the vast majority of Internet users prefer. So for any product, service or information you want people to read about, it helps to create several different versions of your web pages and see what are getting conversions.

    1. It’s amazing the improvements you can get in your conversion rates by testing the smallest things. I recently setup an affiliate site; just by testing 3 different colours in Google Website Optmizer I got 32% improvement in click volume. You really can’t disregard the smallest things in this business.

  14. I love this. Really got my mind going too as far as how I want to build my squeeze pages. I can visualize how my web pages should look more easily now.

    1. I’ve just started with affiliate marketing and I have to admit this post couldn’t have come at a better time. I want to try it though it’s a bit daunting since I have no idea how to start. If find this post and some of the comments I’ve read really motivating.

  15. Split testing is absolutely vital to improving campaigns, but i still believe it is one of the hardest things to track. I have battled with this over the years and still don’t have a totally solid solution for tracking.

    Setting up the split testing is easy, but dont even try it until you have very detailed and accurate tracking of exactly what is working.

    1. Aren’t there tools that can be used to track results? I don’t think I’d need much details. A fairly good idea of what’s generating traffic and conversions would do it for me.

    2. Should I start split testing even if I’ve just launched a site? When will I know if it’s a good time to split test?

  16. Great post – and great real-world examples!

    I’ve always wanted to split test, but what software are you using to really test and prove the improvements?

    What makes you KNOW that the new buttons/header helped – and it wasn’t just the visitors at the time who were more interested in the product?

    How long do you try the new changes? How many do you try?

    Also – do you adjust your adwords bids while doing your split testing? Or do you get these right before doing the split testing?

    Cheers for the post!

    1. I’d like to learn more about this split testing technique. I hope you can do a follow-up post, Justin. Thanks!

  17. It is important to remember though that as with any tests like this, it all depends on the market you are targeting, the product, etc, etc. Just because one button works for one person does not mean it will work for another. If you are serious about increasing your conversions then you should only look at this stuff as advice and test it with your own product yourself. It is silly to assume that this button will increase everyones conversion rates because that just is not true.

    1. You’re right. And from what I understand, he was suggesting to do split testing on buttons – among others. What works or not depends on the result of the testing.

    2. To each his own. But the same principles apply. It’s all a matter of maximizing results that promise to deliver more revenues.

  18. Thanks for sharing. Must start split testing now to increase conversions. More money, more honey!

  19. How long and how often do you do split testing, Justin? In my experience, conversion’s a bit erratic. I should probably start split testing when I notice a dip in traffic and sales. What’s your 2 cents on this?

  20. This sounds like a great way to determine what potential customers are really looking for and what will motivate them to buy products or services. The Internet is such a huge marketplace. Every edge you can find to attract sales is definitely a must-try.

  21. Great post! I’m setting up an online business so this is something that I really should give a try. Looking forward to more of your posts!

  22. I work with affiliates on website design so this information can be very useful to me as well. It’s always good to know what clients need.

  23. For me, Google Website Optimizer is the best free tool available online that can have a massive impact improving profitability.

  24. Makes sense. If you’re looking at potential profit, you’ll have to leave no stone unturned to find what’s providing the most click volume.

  25. Thanks for the informations. I’ve never done such research. And this is really helpful. It makes me wonder that whether i’ve done well to my adsense placement

  26. Though I’m not monetizing my blog at the moment, I plan to in the near future. You have saved me a lot of time and trouble.

  27. I’ve done some testing like this before but never with these results. I think I might have to go try more split testing, thanks.

    1. Online shoppers today have needs and expectations that evolve constantly. With so many to choose from the wide variety of sites and sources to choose from, they have the right to be. It’s not merely a competition of who’s producing the better product or service. It’s all about engaging an online audience that have high expectations.

  28. I can see I need to do some rethinking and some testing of my own. Hope to see more of your useful tips here.

  29. Once I started generating enough revenue from my website I feel like I got lazy on experimenting how to make it better. Even at the start, I haven’t explored split testing though I have heard that it has a huge impact on profitability. About time I give it a shot.

    1. Continuous improvement is key in running any kind of business IMHO. Making small changes is a good start.

  30. I could use this approach when I set up my business’ website. Any more tips you can share on this, Justin?

  31. I learned about this thing a few months back but haven’t bothered to test or experiment with it. It’s high time I try this.

  32. Thanks for sharing this nice piece of information. I’m all for anything that would help with conversions.

  33. Awesome! I’ve been following your posts here especially the insightful series you did. I like that your content’s always clear and doable.

  34. Split Testing is key when it comes to affiliate marketing if your using WordPress there are numerous plug-ins you can use to split test themes such as LP Jumper and A/B split test plug-in. Great article by the way and congrats on the huge profits!

  35. This is a good one. I’ve been planning to make some changes on my website. Seems like a good idea to start with small changes first and split test them to see what works.

  36. Makes me think that adding feature after feature doesn’t guarantee high traffic. That even a small tweak could actually accomplish what complex designs couldn’t . This is great stuff you’ve shared here, Justin! Keep those tips of yours coming!

    1. Most, if not all, of the successful sites I’ve come across have very simple designs. The important thing is to capture your target market’s attention and try hold it long enough for them to make the decision to buy whatever it is you’re selling. Make them easy for them to navigate your site and you’re halfway through getting those conversions you want.

      1. I totally agree with this! I like simple designs myself. If I don’t find what I need within my 2-minute mark then I’m out of that site and off looking for better ones.

  37. Great fat loss program you got there. I used to diet a lot when I gained weight a few years back. Didn’t work out. In the end, I had to rely on eating right and plenty of exercise. I’m curious to know what those secret exercise you have though.

  38. Looks like a great way to measure effectiveness and relevance of content. I hope I’d get my split-testing efforts off to a good start. I’ll have to be selective about what features to tackle first since this may take some work.

    1. When I start to design, I first try to understand what is the important for the people. Because what’s great for me may not be what they’d get them to stay on the page and explore it further.

  39. I’ll be rearranging the steps of my registration process for new customers. That’ll probably have an effect so I’ll start with it.

    1. It’s the getting started part that I find challenging. I’ll have to start with the simplest one I can work on first.

  40. The whole point of split testing is to determine which version of a particular Internet marketing technique yields the best results. So anyone can apply this method to every aspect of a marketing campaign.

  41. Interesting test. The header test I can see why it would increase conversion rates, but the paypal button one really intrigues me.

    I’m wondering if the “Claim Your Copy Now!” part was trigging actions more, have you tried testing that with the original button so they both have that headline right above the button?

  42. Can I use split testing when I create an entirely new site or do I have to wait until such time it has generated a certain amount of traffic?

  43. I’ve read some articles promoting products and services to help make split testing easier and more efficient. But I think Google Website Optimizer is still the best since it’s free.

  44. Conversion rate is a definite multiplier. Interesting how you got a huge jump in conversions just by tweaking buttons and headers.

  45. Any tip you can share on email split-testing? I’d like to find ways to get a better response on my email campaigns.

  46. Sounds awesome. That’s a significant increase in conversions that you got there. Hope I can get the same results too.

  47. The only way find out what will motivate customers is to take action on your offer = measure effectiveness, test your offer, and review results.

  48. I’m surprised that the cartoons made a difference.

    I would have thought that in the weight loss market photos of real people would be seen as evidence of actual results, and that the use of photos would have a negative effect.

    I know photos can be doctored (photoshopped), but basically a cartoon is the ultimate “doctored photo”.

    Go figure!!??!?1?!!

  49. Split Testing is honestly what separates the big boys of internet marketing with the beginners. The guys that make a lot of money simply know how to split test properly to make more and more sales while others are left spitting dust and wondering how they got so good. Right, Justin?

  50. This is perhaps one of the fastest and most effective ways to increase your conversion rate on your website. Nice post as usual, buddy!

    1. I say when you have just one ad out there, it’s difficult to tell what’s working. That’s where split-testing comes in. Split-testing simply means creating a very similar, but ever-so-slightly different ad. If your original ad mentions free shipping at the end of your last line, you might create a mirror ad (as your split test ad) that references the free shipping at the beginning of the last line of your ad.

  51. Is this when you create two or more versions of the page you want to test, each one slightly different than the other? I heard that you “split” your traffic between these two versions and see which one performs better.

  52. So you have an ad that’s up and running. To get to this point you had to find a product, do some keyword research, and write an ad that you think will not only generate clicks, but will also generate sales. That’s great. If you’re doing that, you’re on the right track. But you can even do better by doing a split test.

  53. Split testing is an extremely powerful strategy for increasing the effectiveness of a web based offer. Testing correctly takes time and effort, and the right tools. Got free ones you can suggest?

  54. Simply put, split testing is basically testing two different variables on your sales page to determine which gets the most appeal. “Appeal” meaning sales potential.

  55. I attended a webinar recently where split testing was a topic. I learned that there are four aspects you should really look into: your default sales page, default sales page with a new headline, default sales page with a different price and default sales page with a different background color.

  56. Marketing instinct, experience and intuition can only get you so far. The various factors that make people respond to one thing over another aren’t as well known as the gurus would have you believe. Every product or service is unique, and split testing is the only factor that delivers consistent results.

  57. Identifying an offer on your website that you would like to test is the very first step to a successful split test. What do you think?

  58. I can tell you split testing is the single most important factor in the success of any direct response campaign. I fully support the strategies herein. 😀

  59. Something tells me Shoe’s going to launch another split testing software sooner or later… 😛

    1. Just to see this in another perspective, Justin, what’s the best tool you can recommend?

  60. What variables should I prioritize when I am split testing? An old goat like me hasn’t done it before, to tell you the truth. Thanks in advance! (Don’t hassle the Goff.)

    1. The way it works is you create two or more versions of the page you want to test, each one slightly different than the other. You “split” your traffic between these two versions and see which one performs better.

  61. Another example might be to write an ad without referencing the price of your product. Your second ad could include the price. From there you could track which ad receives a better response. Your thoughts on this?

  62. Any aspect of your sales page such as the color of your headline, the wording of your headline, the images you use on the page, the price of your offer, the text in each body paragraph etc is a variable in split testing.

  63. Instead of just creating a site and hoping that it sells, split testing allows you to increase your conversions over time. Dayuuuuuuuuuum!

  64. A basic split test is designed around testing two variations of a page by delivering the versions equally amongst your visitors and then calculating which version preformed best. Is there an advanced method?

  65. Is it true that there is another type of test called multi-variable testing? I heard it’s a type of testing where you test many variables at once and use software to mathematically figure out the best combination of variables.

  66. One word that you’ll often hear in conjunction with split testing is “statistically significant data.” What does that really mean?

  67. Split testing pros: 1. Simple to implement. 2. Requires little knowledge of statistics. 3. The only approach available to low data rate sites (if you are patient).

    1. Cons? Very small test sizes. Can only test one page element or design at a time. Does not take variable interactions into account.

  68. I prefer multivariate testing myself. You will know EXACTLY what affected your site when you go for it.

    1. You should create a captivating, targeted landing pages anytime you can control where people will arrive on your website from, and your goal is a specific transaction such as a sale, registration, lead-gen, etc.

  69. Thanks for the wonderful report. Now isn’t it time we have another post from Rebecca, too?

  70. he best use of a landing page is not really what it is, but what it can do for you. Your landing pages should provide a personalized sales pitch for the visitor. The best way to do this is consider where the person has originated from, and who they are. By providing a good match, your chance of successfully engaging the visitor increases, as should your conversion rate. This is where split testing comes in.

  71. While it seems that you have been very successful with the splits, I hope you can also give us some tips on how to split test using third party ad services like Adsense. Can you suggest some split testing using them? Thanks

  72. Split testing is very important but I think a change also helps as people get used to the same stuff, hence new buttons work better. Tried the other way? Did conversions decrease?

  73. I don’t believe that just simple buy now button could affect on sales so much. Personally, I would never change my decision of buying a product just because of slightly different buy now button, it just doesn’t makes sense to me, sales letter and site design/impression is what makes buyers (atleast me) to buy a product, not some buttons…

  74. These numbers are useless without more information.

    Was each variation given the same amount of traffic? This is critical information that you didn’t disclose.

  75. This is great info but the title/headline is slightly misleading. Makes it sound like if I split test a useless, no-traffic website I’m all of the sudden going to be raking in the money.

  76. Do you have any ideas as to how you would split test organic traffic? I mean, how would you compare the quality of traffic w/ the type of keywords in relation to main pages of your website?

    What about if the site is database driven like WP or Joomla? I mean, it’s one thing to say “split testing helps” which is useful to people who are just getting online for the first time; but to anyone already making money with larger websites there has to be a more studious approach.

    How would YOU split test a whole design without alienating a large amount of traffic.

  77. I so need to start doing split testing on my landing pages. I’ve always done it when using paid sources like PPC, PPV, and FB… but never really get to in-depth on using it on my landing pages.

    I always feel a sick feeling in the pit of my stomach when I think of the words, ‘A/B Split Test’… but in the end, I’ve heard it THOUSANDS of times… that split testing LP’s pays the bills! -Daniel Wiafe

  78. I’m really amazed by the results you got from your tests. I think everyone ought to pay more attention to details and as a beginner, it’s a trick that I will remember for my future campaigns!!!
    Just a question: How long do you run your tests for?

  79. really useful info, I love the fact Jeremy allows your fantastic guest posts, keep it up!

  80. Not sure how you figured your sales increase of $70,000 when all sales totaled (190 from test #1 and 340 from test #2 @ approx. $40.00 a sale) is $21,200. I must be missing something because I just don’t see the numbers. Anyway, I agree with split testing – it’s a good way to tune your campaign.

  81. That certainly does make split testing seem like its worthwhile , I have just been looking into it, the trouble is I don’t have a buy it now kind of business, just a service business that people phone me to enquire about. Hence I don’t really have a goal page.

  82. I always hear about split testing and frankly it always seems to be too complicated and time-consuming. Thank you Justin for explaining it in a way that makes me interested.
    Split testing sounds alot like sociology and psychology (now I can finally use my degrees! :o)
    Understanding human behavior and what makes a person click. I love it. Thanks again NOW I am going to do it.

  83. Good luck getting people behind this one. Though you make some VERY fascinating points, youre going to have to do more than bring up a few things that may be different than what weve already heard. What are trying to say here? What do you want us to think? It seems like you cant really get behind a unique thought. Anyway, thats just my opinion.

  84. Interesting stuff.. do you reckon it’s the ‘claim your copy now’ above the box or the button image itself that is the main difference..

    Would it also be worth split testing the ‘claim your copy now’ above both graphics with the different buttons?

Comments are closed.