How To Split-Run Test Your Ads

network marketingSuccessful advertising is a combination of art and science. One of the most common mistakes I see among new advertisers is that they seem to think that if they just throw up any old ad, people should begin flocking to their offer. When that doesn’t happen, they blame the advertising service, the company they’re advertising, their business partners, etc… Most of the time, they never really address the actual problem. The problem may very well be that the ad simply isn’t working and needs to be changed.

How can you tell if your ad is working? Ask yourself a few important questions:

Is your ad being placed in front of the right people?
If your company offers nutritional supplements, you want to be sure your ad is being put in front of people that have an interest in nutrition. The local pie eating group may not be the best place to advertise for your offer!

Is your ad being shown repeatedly to your market?
I watch a lot of people run an ad for a day or two, maybe a week. When they don’t see results right away, they give up. Think about the ads that you see over and over again? Why are those ads being repeated? If they weren’t making money for the advertiser, they would be gone. Repetition is a big key to advertising success.

If you can answer “yes” to these questions and your ad still isn’t producing solid responses, then the problem may very well be the ad itself. And this brings me to the point of this article which is:

How To Split-Run Test Your Ads

When it comes to text ads, there are basically 3 components:

  1. Your Subject Line or Headline: This appears at the top of your text ad.
  2. Your Ad Copy: This is where you describe your offer.
  3. Your Call To Action: This is where you tell the reader what to do. For example: Click Below For Free Details.

Each of these 3 components needs to be tested thoroughly so that you can find the perfect combination to excite your prospects to click the ad so they go to your landing page.

What I would do is create 3 ads that have identical ad copy and call to action lines. The only difference between the 3 ads would be the Headline.

Then I would run the ads so that they are all 3 getting a good amount of exposure. You want to run at least a few thousand exposures on each ad before you start trying to analyze the results.

Then check the stats for all 3 ads. The one that has the most clicks is the winner. This tells you which headline is most effective. Let’s say Ad 1 has the winning headline.

Next, I would test the call to action.

Using the SAME 3 ADS, I would start by making sure that they all now have the effective headline. I would then create a new call to action for Ad 2 and Ad 3. (We already know Ad 1 was working better than the other 2).

Depending on your offer and your value proposition, your call to action might be variations of “Click to Learn More!”, “Click For Details”, “Click For Free Details”. “Click for Free Sample”, etc… whatever makes sense for your offer.

After running these new versions for a while, check your results. Again, the ad with the most clicks is the winner and that ad should be left as-is. Let’s say Ad 2 is the winner, this time. So we leave Ad 2 as it is and we change the ad copy for Ad 1 and Ad 3.

The changes may be dramatic or subtle. Run the new ads again. After a few thousands exposures on each one, you should be left with a clear winner.

What next? Delete the ads that aren’t as effective so that all of your exposures can be used on the ad that has proven to be the most effective!

Large marketing firms spend tens of thousands of dollars (or more) on split-run tests before they ever launch the full advertising campaign. With The Online Ad Network, you can run split-run tests for just $19.95 per month. At the end of the process, you will be left with the most effective ad you can create! That’s how successful advertisers get such great results! They are willing to go through the process of testing their ads.


Share on Facebook

42 thoughts on “How To Split-Run Test Your Ads”

  1. Good advice. I have found over the years that changing just one word can make or break an ad’s responsiveness.

    Ideally you should test everything in the ad and continue to test variations forever against the best pulling ad. Keep all your statistics so you can try variations you never tried before. This might sound tedious but the results can often be well worth the trouble.

    • Good point, Bruce.

      I’ve seen HUGE impact by changing a call to action from:

      Call 555.555.5555


      Call NOW 555.555.5555

      Sometimes, it really just is that simple.

  2. Very Interesting post Brian….

    Full of great tips and information to put you on the right track to success.

    Look forward in reading more interesting tips from your blog posts…


    • It depends on what I’m promoting.

      How many click-throughs does it take to get you to your goals?
      That is the only question that any of us need to answer.

      For one offer, a .5% CTR may be fine.
      For another offer, I may need a 2% CTR to achieve the desired goal.

      This is why we teach “test, test, and test some more” so that you can get your ads to perform at the level that helps you achieve your goals for your business.

  3. Hi Brian,

    That’s a very clear explanation of split-testing. Until now I have worked with TWO ads and eliminated the loser. Is there a specific advantage to testing 3 and not 2 ads?



  4. Brian,

    I’m new to your system. I have ads that are cross compatible running in more that one category. The problem I found with your system is that there is no way to see which category is working better for the ad.

    Without allowing me to see the category it is difficult to determine what audience responds better to the ads.

    • Great News! Your stats show the stats for each ad and each ad is displayed separately. You can view the ad details, copy, and category for each ad you create.

  5. Brian,

    Great post I have been with you for about a month I will continue to test. I am getting alot of hits but no customers as of yet I will not give up.

  6. @Brian: You have brought up the fourth item to test: The conversion rate AFTER they click on your ad. Getting ad clicks is first (follow B. Rooney’s advice above to optimize.)

    Then, after you start getting lots of hits, you can start testing your landing page (where you are sending your ads) to see which elements of it work best.

    Testing is an ongoing process for ALL elements of marketing.


    • More good info for me as usual. Starting this tool today. I sure am glad I have condensed my projects down to 4, of which 3 are yours and you are involved in the other one. I have decided if I am ever going to be successful in this business, I am right where I need to be.



  7. I have tested ads (with good results) by personalizing the subject lines to include one of your best offer benefit. Let prospects know what’s in it for them. (example: Robert, Hurry! FREE Unlimited Quality Leads For Your Business!

    • I can certainly see that working when you have their contact info. But with text and banner ads, you don’t really have any way of knowing the names of the people visiting your site.

      But yes … when following up with autoresponders, personalization is a critical key to success.

  8. I have never tested the ads because the ads are many and not knowing how to distinguish one from the other. when others can test their ads why can’t i. Only testing will reveal the true value of the ad.

Leave a Comment